1. POMA
  2. Commit 1b23d900dcbd5240a71883973ac9715537636f10
Browse code

ready to PR devel

pcastellanoescuder authored on 06/12/2023 23:02:30
Showing 45 changed files
.DS_Store
History View file @ 1b23d90
1 1 
new file mode 100644
2 2 
Binary files /dev/null and b/.DS_Store differ
DESCRIPTION
History View file @ 1b23d90
... ... 
@@ -1,6 +1,6 @@
1 1 
 Package: POMA
2 2 
 Title: Tools for Omics Data Analysis
3 
-Version: 1.8.45
3 
+Version: 1.8.47
4 4 
 Authors@R:
5 5 
     c(person(given = "Pol",
6 6 
              family = "Castellano-Escuder",
NAMESPACE
History View file @ 1b23d90
... ... 
@@ -7,7 +7,6 @@ export(PomaCorr)
7 7 
 export(PomaCreateObject)
8 8 
 export(PomaDESeq)
9 9 
 export(PomaDensity)
10 
-export(PomaEDA)
11 10 
 export(PomaHeatmap)
12 11 
 export(PomaImpute)
13 12 
 export(PomaLM)
NEWS.md
History View file @ 1b23d90
... ... 
@@ -1,4 +1,4 @@
1 
-# POMA 1.8.45
1 
+# POMA 1.8.47
2 2
3 3 
 * New POMA theme and colorblind-friendly palette
4 4 
 * Available sample normalization (sum and quantile)
R/PomaEDA.R
History View file @ 1b23d90
5 5 
deleted file mode 100644
... ... 
@@ -1,49 +0,0 @@
1 
-
2 
-#' Automatic Exploratory Data Analysis HTML Report
3 
-#'
4 
-#' @description This function automatically generates a HTML report with different exploratory plots and tables from an `SummarizedExperiment` object.
5 
-#'
6 
-#' @param data A `SummarizedExperiment` object.
7 
-#' @param imputation Imputation method. See `?POMA::PomaImpute()`.
8 
-#' @param normalization Normalization method. See `?POMA::PomaNorm()`.
9 
-#' @param clean_outliers Logical. If it's set to TRUE, outliers will be removed from EDA.
10 
-#' @param coeff_outliers This value corresponds to the classical 1.5 in \eqn{Q3 + 1.5*IQR} formula to detect outliers. See `?POMA::PomaOutliers()`.
11 
-#' @param username Author name in the report.
12 
-#' @param institution Institution name in the report.
13 
-#'
14 
-#' @export
15 
-#'
16 
-#' @return An exploratory data analysis HTML report.
17 
-#' @author Pol Castellano-Escuder
18 
-PomaEDA <- function(data, # nocov start
19 
-                    imputation = "knn",
20 
-                    normalization = "log_pareto",
21 
-                    clean_outliers = TRUE,
22 
-                    coeff_outliers = 1.5,
23 
-                    username = NULL,
24 
-                    institution = NULL){
25 
-
26 
-  .Deprecated(new = return(NULL), msg = "This function has been deprecated")
27 
-
28 
-  if (missing(data)) {
29 
-    stop("data argument is empty!")
30 
-  }
31 
-  if(!is(data, "SummarizedExperiment")){
32 
-    stop("data is not a SummarizedExperiment object. \nSee POMA::PomaSummarizedExperiment or SummarizedExperiment::SummarizedExperiment")
33 
-  }
34 
-  if (!(imputation %in% c("none", "half_min", "median", "mean", "min", "knn"))) {
35 
-    stop("Incorrect value for imputation argument!")
36 
-  }
37 
-  if (!(normalization %in% c("none", "auto_scaling", "level_scaling", "log_scaling",
38 
-                      "log_transformation", "vast_scaling", "log_pareto"))) {
39 
-    stop("Incorrect value for normalization argument!")
40 
-  }
41 
-  if(!is(SummarizedExperiment::colData(data)[,1], "character") &
42 
-     !is(SummarizedExperiment::colData(data)[,1], "factor")){
43 
-    stop("PomaEDA expects the first column of your target to be a factor or character. More report options coming soon...")
44 
-  }
45 
-
46 
-  rmarkdown::render(system.file("rmd", "POMA_EDA_report.Rmd", package = "POMA"), "html_document")
47 
-
48 
-} # nocov end
49 
-
R/PomaLMM.R
History View file @ 1b23d90
... ... 
@@ -48,8 +48,7 @@ PomaLMM <- function(data,
48 48 
                     x = NULL,
49 49 
                     y = NULL,
50 50 
                     adjust = "fdr",
51 
-                    clean_plot = FALSE,
52 
-                    ...) {
51 
+                    clean_plot = FALSE) {
53 52
54 53 
   if (!is(data, "SummarizedExperiment")){
55 54 
     stop("data is not a SummarizedExperiment object. \nSee POMA::PomaCreateObject or SummarizedExperiment::SummarizedExperiment")
R/PomaOddsRatio.R
History View file @ 1b23d90
... ... 
@@ -35,9 +35,9 @@ PomaOddsRatio <- function(data,
35 35 
   }
36 36 
   if (!is.factor(SummarizedExperiment::colData(data)[,1])) {
37 37 
     stop("Grouping factor must be a factor (first column of the metadata file)")
38 
-    if (length(table(SummarizedExperiment::colData(data)[,1])[table(SummarizedExperiment::colData(data)[,1]) != 0]) != 2) {
39 
-      stop("Grouping factor must have exactly 2 levels (first column of the metadata file)")
40 
-    }
38 
+  }
39 
+  if (length(table(SummarizedExperiment::colData(data)[,1])[table(SummarizedExperiment::colData(data)[,1]) != 0]) != 2) {
40 
+    stop("Grouping factor must have exactly 2 levels (first column of the metadata file)")
41 41 
   }
42 42 
   if (!is.null(feature_name)) {
43 43 
     if(!any(feature_name %in% rownames(SummarizedExperiment::assay(data)))) {
... ... 
@@ -99,7 +99,7 @@ PomaOddsRatio <- function(data,
99 99 
     ggplot2::geom_point(size = 3, pch = 21, fill = "orange") +
100 100 
     ggplot2::labs(x = "Odds Ratio",
101 101 
                   y = NULL) +
102 
-    POMA::theme_poma()
102 
+    theme_poma()
103 103
104 104 
   return(list(odds_ratio_table = odds,
105 105 
               odds_ratio_plot = ORPlot))
R/PomaPLS.R
History View file @ 1b23d90
... ... 
@@ -324,7 +324,7 @@ PomaPLS <- function(data,
324 324 
                                            test.keepX = c(1:num_features), nrepeat = nrepeat)
325 325
326 326 
       opt_ncomp <- tune_splsda$choice.ncomp$ncomp # optimal number of components based on t-tests
327 
-      select_keepX <- tune_splsda$choice.keepX[1:ncomp]  # optimal number of variables to select
327 
+      select_keepX <- tune_splsda$choice.keepX[1:ncomp] # optimal number of variables to select
328 328
329 329 
       errors_splsda_sd <- data.frame(tune_splsda$error.rate.sd) %>%
330 330 
         tibble::rownames_to_column("feature_sd") %>%
R/PomaRankProd.R
History View file @ 1b23d90
... ... 
@@ -75,12 +75,8 @@ PomaRankProd <- function(data,
75 75
76 76 
   one <- as.data.frame(top_rank$Table1)
77 77 
   two <- as.data.frame(top_rank$Table2)
78 
-
79 
-  if(nrow(one) == 0 & nrow(two) == 0){
80 
-    stop("No significant features found...")
81 
-  }
82 78
83 
-  if(nrow(one) != 0){
79 
+  if (nrow(one) != 0){
84 80
85 81 
     one <- one %>%
86 82 
       tibble::rownames_to_column("feature") %>%
... ... 
@@ -92,7 +88,7 @@ PomaRankProd <- function(data,
92 88 
     colnames(one)[4] <- paste0("FC_", class1, "_", class2)
93 89 
   }
94 90
95 
-  if(nrow(two) != 0){
91 
+  if (nrow(two) != 0){
96 92
97 93 
     two <- two %>%
98 94 
       tibble::rownames_to_column("feature") %>%
... ... 
@@ -132,14 +128,14 @@ PomaRankProd <- function(data,
132 128
133 129 
   plot1 <- ggplot2::ggplot(rp_plot, ggplot2::aes(x = rank1, y = pfp1)) +
134 130 
     ggplot2::geom_point(size = 1.5, alpha=0.9) +
135 
-    ggplot2::theme_bw() +
131 
+    theme_poma() +
136 132 
     ggplot2::labs(x = "Number of identified features",
137 133 
                   y = "Estimated PFP",
138 134 
                   title = paste0("Up-regulated features in ", class2))
139 135
140 136 
   plot2 <- ggplot2::ggplot(rp_plot, ggplot2::aes(x = rank2, y = pfp2)) +
141 137 
     ggplot2::geom_point(size = 1.5, alpha=0.9) +
142 
-    ggplot2::theme_bw() +
138 
+    theme_poma() +
143 139 
     ggplot2::labs(x = "Number of identified features",
144 140 
                   y = "Estimated PFP",
145 141 
                   title = paste0("Down-regulated features in ", class2))
... ... 
@@ -148,6 +144,5 @@ PomaRankProd <- function(data,
148 144 
               down_regulated = two,
149 145 
               up_regulated_plot = plot1,
150 146 
               down_regulated_plot = plot2))
151 
-
152 147 
 }
153 148
R/PomaUMAP.R
History View file @ 1b23d90
... ... 
@@ -48,6 +48,8 @@ PomaUMAP <- function(data,
48 48 
     stop("data is not a SummarizedExperiment object. \nSee POMA::PomaCreateObject or SummarizedExperiment::SummarizedExperiment")
49 49 
   }
50 50
51 
+  if (hdbscan_minpts < 2) {hdbscan_minpts <- 2}
52 
+
51 53 
   to_umap <- t(SummarizedExperiment::assay(data))
52 54
53 55 
   umap_res <- uwot::umap(to_umap,
R/PomaVolcano.R
History View file @ 1b23d90
... ... 
@@ -42,8 +42,7 @@ PomaVolcano <- function(data,
42 42 
                         log2fc_cutoff = NULL,
43 43 
                         labels = FALSE,
44 44 
                         paired = FALSE,
45 
-                        var_equal = FALSE,
46 
-                        ...) {
45 
+                        var_equal = FALSE) {
47 46
48 47 
   if(!is(data, "SummarizedExperiment")){
49 48 
     stop("data is not a SummarizedExperiment object. \nSee POMA::PomaCreateObject or SummarizedExperiment::SummarizedExperiment")
R/helpers.R
History View file @ 1b23d90
... ... 
@@ -43,12 +43,30 @@ make_legend <- function(fun,
43 43 
   return(legend)
44 44 
 }
45 45
46 
-create_mock_summarized_experiment <- function(binary = FALSE) {
46 
+create_mock_summarized_experiment <- function(binary = FALSE, paired = FALSE, integers = FALSE) {
47 47
48 
-  if (!binary) {g_labels <- c("A", "B", "C")} else {g_labels <- c("A", "B")}
48 
+  if (!binary) {
49 
+    g_labels <- sample(c("A", "B", "C"), 20, replace = TRUE)
50 
+  } else {
51 
+    g_labels <- sample(c("A", "B"), 20, replace = TRUE)
52 
+    if (paired) {
53 
+      g_labels <- c(rep("A", 10), rep("B", 10))
54 
+    }
55 
+  }
56 
+
57 
+  if (integers) {
58 
+    matrix_data <- matrix(sample(1:100, 20 * 10, replace = TRUE), nrow = 20, ncol = 10)
59 
+  } else {
60 
+    matrix_data <- matrix(runif(100), nrow = 20)
61 
+  }
49 62
50 
-  matrix_data <- matrix(runif(100), nrow = 20)
51 
-  col_data <- data.frame(sample = paste0("Sample", 1:20), group = sample(g_labels, 20, replace = TRUE))
63 
+  col_data <- data.frame(sample = paste0("Sample", 1:20), group = g_labels)
52 64 
   PomaCreateObject(features = matrix_data, metadata = col_data)
53 65 
 }
54 66
67 
+create_mock_data <- function() {
68 
+  features <- as.data.frame(matrix(runif(100), ncol = 10))
69 
+  metadata <- data.frame(ID = 1:10, Group = factor(rep(c("A", "B"), each = 5)))
70 
+  list(features = features, metadata = metadata)
71 
+}
72 
+
README.Rmd
History View file @ 1b23d90
... ... 
@@ -16,7 +16,7 @@ output: github_document
16 16 
 | _BioC_ branch 	| Status 	| Version 	| Dependencies 	| Rank 	|
17 17 
 |-	|-	|-	|-	|-	|
18 18 
 | [Release](http://bioconductor.org/packages/release/bioc/html/POMA.html) 	| [![Bioc release status](https://bioconductor.org/shields/build/release/bioc/POMA.svg)](https://bioconductor.org/checkResults/release/bioc-LATEST/POMA/) 	| [![BioC released version](https://img.shields.io/badge/release%20version-1.6.0-blue.svg)](https://www.bioconductor.org/packages/POMA) 	| [![Dependencies](http://bioconductor.org/shields/dependencies/release/POMA.svg)](http://bioconductor.org/packages/release/bioc/html/POMA.html#since) 	| [![Rank](http://www.bioconductor.org/shields/downloads/release/POMA.svg)](https://bioconductor.org/packages/stats/bioc/POMA) 	|
19 
-| [Devel](http://bioconductor.org/packages/devel/bioc/html/POMA.html) 	| [![Bioc devel status](https://bioconductor.org/shields/build/devel/bioc/POMA.svg)](https://bioconductor.org/checkResults/devel/bioc-LATEST/POMA/) 	| [![BioC devel version](https://img.shields.io/badge/devel%20version-1.8.45-blue.svg)](https://bioconductor.org/packages/devel/bioc/html/POMA.html) 	| [![Dependencies](http://bioconductor.org/shields/dependencies/devel/POMA.svg)](http://bioconductor.org/packages/devel/bioc/html/POMA.html#since) 	| [![Rank](http://www.bioconductor.org/shields/downloads/devel/POMA.svg)](https://bioconductor.org/packages/stats/bioc/POMA) 	|
19 
+| [Devel](http://bioconductor.org/packages/devel/bioc/html/POMA.html) 	| [![Bioc devel status](https://bioconductor.org/shields/build/devel/bioc/POMA.svg)](https://bioconductor.org/checkResults/devel/bioc-LATEST/POMA/) 	| [![BioC devel version](https://img.shields.io/badge/devel%20version-1.8.47-blue.svg)](https://bioconductor.org/packages/devel/bioc/html/POMA.html) 	| [![Dependencies](http://bioconductor.org/shields/dependencies/devel/POMA.svg)](http://bioconductor.org/packages/devel/bioc/html/POMA.html#since) 	| [![Rank](http://www.bioconductor.org/shields/downloads/devel/POMA.svg)](https://bioconductor.org/packages/stats/bioc/POMA) 	|
20 20
21 21 
   <!-- badges: end -->
22 22
README.md
History View file @ 1b23d90
... ... 
@@ -18,7 +18,7 @@ v3](https://img.shields.io/badge/License-GPLv3-blue.svg)](https://www.gnu.org/li
18 18 
 | *BioC* branch                                                           | Status                                                                                                                                                  | Version                                                                                                                                            | Dependencies                                                                                                                                         | Rank                                                                                                                         |
19 19 
 |-------------------------------------------------------------------------|---------------------------------------------------------------------------------------------------------------------------------------------------------|----------------------------------------------------------------------------------------------------------------------------------------------------|------------------------------------------------------------------------------------------------------------------------------------------------------|------------------------------------------------------------------------------------------------------------------------------|
20 20 
 | [Release](http://bioconductor.org/packages/release/bioc/html/POMA.html) | [![Bioc release status](https://bioconductor.org/shields/build/release/bioc/POMA.svg)](https://bioconductor.org/checkResults/release/bioc-LATEST/POMA/) | [![BioC released version](https://img.shields.io/badge/release%20version-1.6.0-blue.svg)](https://www.bioconductor.org/packages/POMA)              | [![Dependencies](http://bioconductor.org/shields/dependencies/release/POMA.svg)](http://bioconductor.org/packages/release/bioc/html/POMA.html#since) | [![Rank](http://www.bioconductor.org/shields/downloads/release/POMA.svg)](https://bioconductor.org/packages/stats/bioc/POMA) |
21 
-| [Devel](http://bioconductor.org/packages/devel/bioc/html/POMA.html)     | [![Bioc devel status](https://bioconductor.org/shields/build/devel/bioc/POMA.svg)](https://bioconductor.org/checkResults/devel/bioc-LATEST/POMA/)       | [![BioC devel version](https://img.shields.io/badge/devel%20version-1.8.45-blue.svg)](https://bioconductor.org/packages/devel/bioc/html/POMA.html) | [![Dependencies](http://bioconductor.org/shields/dependencies/devel/POMA.svg)](http://bioconductor.org/packages/devel/bioc/html/POMA.html#since)     | [![Rank](http://www.bioconductor.org/shields/downloads/devel/POMA.svg)](https://bioconductor.org/packages/stats/bioc/POMA)   |
21 
+| [Devel](http://bioconductor.org/packages/devel/bioc/html/POMA.html)     | [![Bioc devel status](https://bioconductor.org/shields/build/devel/bioc/POMA.svg)](https://bioconductor.org/checkResults/devel/bioc-LATEST/POMA/)       | [![BioC devel version](https://img.shields.io/badge/devel%20version-1.8.47-blue.svg)](https://bioconductor.org/packages/devel/bioc/html/POMA.html) | [![Dependencies](http://bioconductor.org/shields/dependencies/devel/POMA.svg)](http://bioconductor.org/packages/devel/bioc/html/POMA.html#since)     | [![Rank](http://www.bioconductor.org/shields/downloads/devel/POMA.svg)](https://bioconductor.org/packages/stats/bioc/POMA)   |
22 22
23 23 
 <!-- badges: end -->
24 24
inst/rmd/POMA_EDA_report.Rmd
History View file @ 1b23d90
25 25 
deleted file mode 100644
... ... 
@@ -1,239 +0,0 @@
1 
-title: "Exploratory Data Analysis Report"
2 
-subtitle: "Generated with POMA `r paste0('(', packageVersion('POMA'), ')')`"
3 
-author: '`r paste0(username, institution)`'
4 
-date: '`r Sys.Date()`'
5 
-output:
6 
-  html_document:
7 
-    toc: true
8 
-    number_sections: true
9 
-
10 
-```{r, echo = FALSE, warning = FALSE, comment = NA, message = FALSE}
11 
-# This file is part of POMA.
12 
-
13 
-# POMA is free software: you can redistribute it and/or modify
14 
-# it under the terms of the GNU General Public License as published by
15 
-# the Free Software Foundation, either version 3 of the License, or
16 
-# (at your option) any later version.
17 
-
18 
-# POMA is distributed in the hope that it will be useful,
19 
-# but WITHOUT ANY WARRANTY; without even the implied warranty of
20 
-# MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the
21 
-# GNU General Public License for more details.
22 
-
23 
-# You should have received a copy of the GNU General Public License
24 
-# along with POMA. If not, see <https://www.gnu.org/licenses/>.
25 
-
26 
-library(POMA)
27 
-
28 
-e <- t(SummarizedExperiment::assay(data))
29 
-target <- SummarizedExperiment::colData(data) %>%
30 
-  as.data.frame() %>%
31 
-  tibble::rownames_to_column("ID") %>%
32 
-  dplyr::rename(Group = 2) %>%
33 
-  dplyr::select(ID, Group)
34 
-```
35 
-
36 
-```{r, echo = FALSE, warning = FALSE, comment = NA, message = FALSE}
37 
-if(clean_outliers){
38 
-  imputed <- PomaImpute(data, method = imputation)
39 
-  pre_processed <- PomaNorm(imputed, method = normalization) %>%
40 
-    PomaOutliers(coef = coeff_outliers)
41 
-} else {
42 
-  imputed <- PomaImpute(data, method = imputation)
43 
-  pre_processed <- PomaNorm(imputed, method = normalization)
44 
-}
45 
-```
46 
-
47 
-```{r, echo = FALSE, warning = FALSE, comment = NA, message = FALSE}
48 
-# zeros
49 
-zeros <- data.frame(number = colSums(e == 0, na.rm = TRUE)) %>%
50 
-  tibble::rownames_to_column("names") %>%
51 
-  dplyr::filter(number != 0)
52 
-
53 
-all_zero <- zeros %>%
54 
-  dplyr::filter(number == nrow(e))
55 
-
56 
-# missing values
57 
-nas <- data.frame(number = colSums(is.na(e))) %>%
58 
-  tibble::rownames_to_column("names") %>%
59 
-  dplyr::filter(number != 0)
60 
-
61 
-# zero variance
62 
-var_zero <- e %>%
63 
-  as.data.frame() %>%
64 
-  dplyr::summarise_all(~ var(., na.rm = TRUE)) %>%
65 
-  t() %>%
66 
-  as.data.frame() %>%
67 
-  tibble::rownames_to_column("names") %>%
68 
-  dplyr::filter(V1 == 0)
69 
-```
70 
-
71 
-# Know your data
72 
-
73 
-  + Your data have **`r nrow(e)`** samples, **`r ncol(e)`** features and **`r length(table(target$Group))`** groups, that are **`r noquote(paste(shQuote(levels(as.factor(target$Group))), collapse=", "))`**. `r ifelse(ncol(SummarizedExperiment::colData(data)) > 1, paste0("Furthermore, **", ncol(SummarizedExperiment::colData(data)) - 1,"** covariates have been found in your data. These covariates are **",noquote(paste(shQuote(paste0(colnames(SummarizedExperiment::colData(data))[2:ncol(SummarizedExperiment::colData(data))])), collapse=", ")),"**."), "")`
74 
-
75 
-  + A **`r round((sum(is.na(e))/(nrow(e)*ncol(e)))*100, 2)`%** of values in your data are NAs (missing values). `r ifelse(nrow(nas) >= 1, paste0("Variables that have NA values are **",noquote(paste(shQuote(paste0(nas$names," (",nas$number,")")), collapse=", ")),"**."), "")`
76 
-
77 
-  + A **`r round((sum(zeros$number)/(nrow(e)*ncol(e)))*100, 2)`%** of values in your data are zeros. `r ifelse(nrow(zeros) >= 1, paste0("Variables that have zeros are **",noquote(paste(shQuote(paste0(zeros$names," (",zeros$number,")")), collapse=", ")),"**."), "")`
78 
-
79 
-  + Removed from the exploratory data analysis **`r nrow(all_zero)`** features that only have zeros. `r ifelse(nrow(all_zero) >= 1, paste0("These variables are **",noquote(paste(shQuote(all_zero$names), collapse=", ")),"**."), "")`
80 
-
81 
-  + Removed from the exploratory data analysis **`r nrow(var_zero)`** features that have zero variance. `r ifelse(nrow(var_zero) >= 1, paste0("These variables are **",noquote(paste(shQuote(var_zero$names), collapse=", ")),"**."), "")`
82 
-
83 
-## Summary Tables
84 
-
85 
-```{r, echo = FALSE, warning = FALSE, comment = NA, message = FALSE}
86 
-summary_table1 <- data.frame(Samples = nrow(e),
87 
-                             Features = ncol(e),
88 
-                             Covariates = ncol(SummarizedExperiment::colData(data)) - 1)
89 
-
90 
-summary_table2 <- data.frame(Number_Zeros = sum(zeros$number),
91 
-                             Percentage_Zeros = paste(round((sum(zeros$number)/(nrow(e)*ncol(e)))*100, 2), "%"))
92 
-
93 
-summary_table3 <- data.frame(Number_Missings = sum(is.na(e)),
94 
-                             Percentage_Missings = paste(round((sum(is.na(e))/(nrow(e)*ncol(e)))*100, 2), "%"))
95 
-
96 
-summary_table1
97 
-summary_table2
98 
-summary_table3
99 
-```
100 
-
101 
-```{r, echo = FALSE, warning = FALSE, comment = NA, message = FALSE, dpi = 300}
102 
-if (nrow(nas) >= 1){
103 
-  ggplot2::ggplot(nas, ggplot2::aes(reorder(names, number), number, fill = number)) +
104 
-    ggplot2::geom_col() +
105 
-    ggplot2::labs(x = NULL,
106 
-                  y = "Missing values",
107 
-                  title = "Missing Value Plot") +
108 
-    ggplot2::theme_bw() +
109 
-    ggplot2::theme(axis.text.x = ggplot2::element_text(angle = 45, hjust = 1),
110 
-                   legend.position = "none") +
111 
-    ggplot2::scale_fill_viridis_c(begin = 0, end = 0.8)
112 
-}
113 
-```
114 
-
115 
-```{r, echo = FALSE, warning = FALSE, comment = NA, message = FALSE, dpi = 300}
116 
-if (nrow(zeros) >= 1){
117 
-  ggplot2::ggplot(zeros, ggplot2::aes(reorder(names, number), number, fill = number)) +
118 
-    ggplot2::geom_col() +
119 
-    ggplot2::labs(x = NULL,
120 
-              y = "Zeros",
121 
-              title = "Zeros Plot") +
122 
-    ggplot2::theme_bw() +
123 
-    ggplot2::theme(axis.text.x = ggplot2::element_text(angle = 45, hjust = 1),
124 
-                   legend.position = "none") +
125 
-    ggplot2::scale_fill_viridis_c(begin = 0, end = 0.8)
126 
-}
127 
-```
128 
-
129 
-## Samples by Group
130 
-
131 
-```{r, echo = FALSE, warning = FALSE, comment = NA, message = FALSE, dpi = 300}
132 
-counts <- data.frame(table(target$Group))
133 
-colnames(counts) <- c("Group", "Counts")
134 
-
135 
-ggplot2::ggplot(counts, ggplot2::aes(reorder(Group, Counts), Counts, fill = Group)) +
136 
-  ggplot2::geom_col() +
137 
-  ggplot2::labs(x = NULL,
138 
-                y = "Counts") +
139 
-  ggplot2::theme_bw() +
140 
-  ggplot2::theme(axis.text.x = ggplot2::element_text(angle = 45, hjust = 1),
141 
-                 legend.position = "none") +
142 
-  ggplot2::scale_fill_viridis_d(begin = 0, end = 0.8)
143 
-```
144 
-
145 
-# Normalization Plots
146 
-
147 
-```{r, echo = FALSE, warning = FALSE, comment = NA, message = FALSE, dpi = 300}
148 
-indNum <- nrow(SummarizedExperiment::colData(pre_processed))
149 
-jttr <- ifelse(indNum <= 10, TRUE, FALSE)
150 
-
151 
-p1 <- PomaBoxplots(imputed,
152 
-                   jitter = jttr,
153 
-                   label_size = 8,
154 
-                   legend_position = "bottom") +
155 
-  ggplot2::labs(x = "Samples",
156 
-                y = "Value",
157 
-                title = "Not Normalized")
158 
-
159 
-p2 <- PomaBoxplots(pre_processed,
160 
-                   jitter = jttr,
161 
-                   label_size = 8,
162 
-                   legend_position = "bottom") +
163 
-  ggplot2::labs(x = "Samples",
164 
-                y = "Value",
165 
-                title = paste0("Normalized (", normalization, ")"))
166 
-
167 
-p1
168 
-p2
169 
-```
170 
-
171 
-# Group Distribution Plots
172 
-
173 
-```{r, echo = FALSE, warning = FALSE, comment = NA, message = FALSE, dpi = 300}
174 
-p3 <- PomaDensity(imputed) +
175 
-  ggplot2::ggtitle("Not Normalized")
176 
-
177 
-p4 <- PomaDensity(pre_processed) +
178 
-  ggplot2::ggtitle(paste0("Normalized (", normalization, ")"))
179 
-
180 
-p3
181 
-p4
182 
-```
183 
-
184 
-# Outlier Detection
185 
-
186 
-```{r, echo = FALSE, warning = FALSE, comment = NA, message = FALSE}
187 
-outliers <- data %>%
188 
-  PomaImpute(method = imputation) %>%
189 
-  PomaNorm(method = normalization) %>%
190 
-  PomaOutliers(do = "analyze", coef = coeff_outliers)
191 
-outliers$polygon_plot
192 
-```
193 
-
194 
-**`r nrow(outliers$outliers)`** possible outliers detected in your data. `r ifelse(nrow(outliers$outliers) >= 1, paste0("These outliers are **",noquote(paste(shQuote(paste0(outliers$outliers$sample)), collapse=", ")),"**."), "")`
195 
-
196 
-```{r, echo = FALSE, warning = FALSE, comment = NA, message = FALSE}
197 
-if(nrow(outliers$outliers) >= 1){
198 
-  outliers$outliers
199 
-  }
200 
-```
201 
-
202 
-# High Correlated Features (r > 0.97)
203 
-
204 
-```{r, echo = FALSE, warning = FALSE, comment = NA, message = FALSE}
205 
-correlations <- PomaCorr(pre_processed,
206 
-                         label_size = 8)
207 
-
208 
-high_correlations <- correlations$correlations %>%
209 
-  dplyr::filter(abs(corr) > 0.97)
210 
-```
211 
-
212 
-There are **`r nrow(high_correlations)`** high correlated feature pairs in your data. `r ifelse(nrow(high_correlations) >= 1, paste0("These features are **",noquote(paste(shQuote(paste0(high_correlations$feature1, " - " , high_correlations$feature2)), collapse=", ")),"**."), "")`
213 
-
214 
-```{r, echo = FALSE, warning = FALSE, comment = NA, message = FALSE, dpi = 300, fig.align = 'center'}
215 
-correlations$corrplot
216 
-```
217 
-
218 
-# Heatmap and Clustering
219 
-
220 
-```{r, echo = FALSE, warning = FALSE, comment = NA, message = FALSE, dpi = 300}
221 
-PomaHeatmap(pre_processed, sample_names = FALSE)
222 
-```
223 
-
224 
-# Principal Component Analysis
225 
-
226 
-```{r, echo = FALSE, warning = FALSE, comment = NA, message = FALSE, dpi = 300}
227 
-PomaMultivariate(pre_processed, method = "pca", ellipse = FALSE)$scoresplot
228 
-```
229 
-
230 
-# Uniform Manifold Approximation and Projection Clustering
231 
-
232 
-```{r, echo = FALSE, warning = FALSE, comment = NA, message = FALSE, dpi = 300}
233 
-PomaUMAP(pre_processed,
234 
-         hdbscan_minpts = 5,
235 
-         show_clusters = TRUE)$umap_plot
236 
-```
237 
-
man/PomaEDA.Rd
History View file @ 1b23d90
238 0 
deleted file mode 100644
... ... 
@@ -1,40 +0,0 @@
1 
-% Generated by roxygen2: do not edit by hand
2 
-% Please edit documentation in R/PomaEDA.R
3 
-\name{PomaEDA}
4 
-\alias{PomaEDA}
5 
-\title{Automatic Exploratory Data Analysis HTML Report}
6 
-\usage{
7 
-PomaEDA(
8 
-  data,
9 
-  imputation = "knn",
10 
-  normalization = "log_pareto",
11 
-  clean_outliers = TRUE,
12 
-  coeff_outliers = 1.5,
13 
-  username = NULL,
14 
-  institution = NULL
15 
-)
16 
-}
17 
-\arguments{
18 
-\item{data}{A \code{SummarizedExperiment} object.}
19 
-
20 
-\item{imputation}{Imputation method. See \code{?POMA::PomaImpute()}.}
21 
-
22 
-\item{normalization}{Normalization method. See \code{?POMA::PomaNorm()}.}
23 
-
24 
-\item{clean_outliers}{Logical. If it's set to TRUE, outliers will be removed from EDA.}
25 
-
26 
-\item{coeff_outliers}{This value corresponds to the classical 1.5 in \eqn{Q3 + 1.5*IQR} formula to detect outliers. See \code{?POMA::PomaOutliers()}.}
27 
-
28 
-\item{username}{Author name in the report.}
29 
-
30 
-\item{institution}{Institution name in the report.}
31 
-}
32 
-\value{
33 
-An exploratory data analysis HTML report.
34 
-}
35 
-\description{
36 
-This function automatically generates a HTML report with different exploratory plots and tables from an \code{SummarizedExperiment} object.
37 
-}
38 
-\author{
39 
-Pol Castellano-Escuder
40 
-}
man/PomaLMM.Rd
History View file @ 1b23d90
... ... 
@@ -4,7 +4,7 @@
4 4 
 \alias{PomaLMM}
5 5 
 \title{Linear Mixed Models}
6 6 
 \usage{
7 
-PomaLMM(data, x = NULL, y = NULL, adjust = "fdr", clean_plot = FALSE, ...)
7 
+PomaLMM(data, x = NULL, y = NULL, adjust = "fdr", clean_plot = FALSE)
8 8 
 }
9 9 
 \arguments{
10 10 
 \item{data}{A \code{SummarizedExperiment} object.}
man/PomaVolcano.Rd
History View file @ 1b23d90
... ... 
@@ -13,8 +13,7 @@ PomaVolcano(
13 13 
   log2fc_cutoff = NULL,
14 14 
   labels = FALSE,
15 15 
   paired = FALSE,
16 
-  var_equal = FALSE,
17 
-  ...
16 
+  var_equal = FALSE
18 17 
 )
19 18 
 }
20 19 
 \arguments{
tests/.DS_Store
History View file @ 1b23d90
21 20 
new file mode 100644
22 21 
Binary files /dev/null and b/tests/.DS_Store differ
tests/testthat/.DS_Store
History View file @ 1b23d90
23 22 
new file mode 100644
24 23 
Binary files /dev/null and b/tests/testthat/.DS_Store differ
tests/testthat/test-PomaBoxplots.R
History View file @ 1b23d90
... ... 
@@ -1,68 +1,42 @@
1 
-context("PomaBoxplots")
2 1
3 
-test_that("PomaBoxplots works", {
4 
-
5 
-  data("st000284")
6 
-
7 
-  norm_none <- PomaNorm(st000284, method = "none")
8 
-  norm_ls <- PomaNorm(st000284, method = "log_scaling")
9 
-
10 
-  a <- PomaBoxplots(norm_none, label_size = 12)
11 
-  b <- PomaBoxplots(norm_ls, label_size = 10)
12 
-  c <- PomaBoxplots(norm_none, x = "features")
13 
-  d <- PomaBoxplots(norm_ls, x = "features")
14 
-
15 
-  e <- PomaBoxplots(norm_none, x = "samples")
16 
-
17 
-  f <- PomaBoxplots(norm_none, x = "samples", violin = TRUE)
18 
-  g <- PomaBoxplots(norm_none, x = "samples", violin = FALSE)
19 
-  h <- PomaBoxplots(norm_none, x = "features", violin = TRUE)
20 
-  i <- PomaBoxplots(norm_none, x = "features", violin = FALSE)
21 
-
22 
-  j <- PomaBoxplots(norm_ls, x = "features", feature_name = "methyl_succinate")
23 
-  k <- PomaBoxplots(norm_ls, x = "features", feature_name = c("methyl_succinate", "linoleic_acid"))
24 
-
25 
-
26 
-  df_a <- ggplot2::layer_data(a)
27 
-  df_b <- ggplot2::layer_data(b)
28 
-  df_c <- ggplot2::layer_data(c)
29 
-  df_d <- ggplot2::layer_data(d)
30 
-  df_e <- ggplot2::layer_data(e)
31 
-
32 
-  df_f <- ggplot2::layer_data(f)
33 
-  df_g <- ggplot2::layer_data(g)
34 
-  df_h <- ggplot2::layer_data(h)
35 
-  df_i <- ggplot2::layer_data(i)
36 
-
37 
-  df_j <- ggplot2::layer_data(j)
38 
-  df_k <- ggplot2::layer_data(k)
39 
-
40 
-  ####
41 
-
42 
-  expect_true(min(df_a$ymin) == min(df_c$ymin))
43 
-  expect_true(min(df_b$ymin) != min(df_d$ymin))
44 
-  expect_false(all(df_a$ymin == df_b$ymin))
45 
-  expect_false(all(df_c$ymin == df_d$ymin))
46 
-
47 
-  expect_equal(df_a, df_e)
48 
-
49 
-  expect_false(nrow(df_j) == nrow(df_k))
50 
-  expect_false(nrow(df_j) == nrow(df_h))
51 
-  expect_false(nrow(df_k) == nrow(df_i))
52 
-
53 
-  ##
54 
-
55 
-  expect_error(PomaBoxplots(norm_ls, x = "samp"))
56 
-  expect_error(PomaBoxplots(x = "sample"))
57 
-  expect_error(PomaBoxplots(iris, x = "sample"))
58 
-
59 
-  ##
60 
-
61 
-  expect_error(PomaBoxplots(norm_ls, x = "features", feature_name = "hello"))
62 
-  expect_error(PomaBoxplots(norm_ls, x = "features", feature_name = "methyl_succina"))
63 
-
64 
-  expect_message(PomaBoxplots(norm_ls, x = "features", feature_name = c("methyl_succina", "linoleic_acid")))
65 
-  expect_message(PomaBoxplots(norm_ls, x = "features", feature_name = c("methyl_succinate", "linoleic_aci")))
66 
-
2 
+test_that("PomaBoxplots handles valid SummarizedExperiment objects", {
3 
+  data <- create_mock_summarized_experiment()
4 
+  plot_samples <- PomaBoxplots(data, x = "samples")
5 
+  plot_features <- PomaBoxplots(data, x = "features")
6 
+  expect_is(plot_samples, "ggplot")
7 
+  expect_is(plot_features, "ggplot")
8 
+})
9 
+
10 
+test_that("PomaBoxplots stops with non-SummarizedExperiment objects", {
11 
+  data <- data.frame(matrix(runif(100), ncol = 10))
12 
+  expect_error(PomaBoxplots(data), "data is not a SummarizedExperiment object")
13 
+})
14 
+
15 
+test_that("PomaBoxplots handles violin plot option correctly", {
16 
+  data <- create_mock_summarized_experiment()
17 
+  plot_violin <- PomaBoxplots(data, violin = TRUE)
18 
+  expect_is(plot_violin, "ggplot")
19 
+})
20 
+
21 
+test_that("PomaBoxplots stops with incorrect x argument", {
22 
+  data <- create_mock_summarized_experiment()
23 
+  expect_error(PomaBoxplots(data, x = "invalid_option"), "Incorrect value for x argument")
24 
+})
25 
+
26 
+test_that("PomaBoxplots handles feature_name parameter correctly", {
27 
+  data <- create_mock_summarized_experiment()
28 
+  plot_specific_feature <- PomaBoxplots(data, x = "features", feature_name = c("V2"))
29 
+  expect_is(plot_specific_feature, "ggplot")
30 
+})
31 
+
32 
+test_that("PomaBoxplots stops with non-existing feature names", {
33 
+  data <- create_mock_summarized_experiment()
34 
+  expect_error(PomaBoxplots(data, x = "features", feature_name = c("non_existing_feature")), "Features not found")
35 
+})
36 
+
37 
+test_that("PomaBoxplots applies theme parameters correctly", {
38 
+  data <- create_mock_summarized_experiment()
39 
+  plot_with_theme <- PomaBoxplots(data, theme_params = list(legend_title = TRUE))
40 
+  expect_is(plot_with_theme, "ggplot")
67 41 
 })
68 42
tests/testthat/test-PomaClust.R
History View file @ 1b23d90
... ... 
@@ -1,64 +1,34 @@
1 
-context("PomaClust")
2 1
3 
-test_that("PomaClust works", {
4 
-
5 
-  data("st000284")
6 
-  data("st000336")
7 
-  imp_st000336 <- PomaImpute(st000336, method = "knn")
8 
-
9 
-  a <- PomaClust(st000284)
10 
-  b <- PomaClust(imp_st000336)
11 
-
12 
-  c <- PomaClust(st000284, method = "maximum", k = 5, show_clusters = FALSE, labels = TRUE)
13 
-  d <- PomaClust(imp_st000336, method = "manhattan", k = 2, show_clusters = FALSE, labels = TRUE)
14 
-
15 
-  e <- PomaClust(st000284, method = "canberra", k = 6, show_clusters = FALSE, labels = TRUE, show_group = TRUE)
16 
-  f <- PomaClust(imp_st000336, method = "minkowski", k = 4, show_clusters = TRUE, labels = TRUE, show_group = TRUE)
17 
-
18 
-  g <- PomaClust(st000284)
19 
-  h <- PomaClust(st000284, k = 3)
20 
-
21 
-  ## table
22 
-
23 
-  expect_equal(nrow(a$mds_values), nrow(c$mds_values))
24 
-  expect_equal(nrow(b$mds_values), nrow(d$mds_values))
25 
-  expect_equal(nrow(e$mds_values), nrow(a$mds_values))
26 
-  expect_equal(nrow(f$mds_values), nrow(b$mds_values))
27 
-
28 
-  expect_equal(5, ncol(a$mds_values))
29 
-  expect_equal(5, ncol(b$mds_values))
30 
-  expect_equal(5, ncol(c$mds_values))
31 
-  expect_equal(5, ncol(d$mds_values))
32 
-  expect_equal(5, ncol(e$mds_values))
33 
-  expect_equal(5, ncol(f$mds_values))
34 
-
35 
-  ## plot
36 
-
37 
-  expect_equal(class(a$mds_plot)[2], "ggplot")
38 
-  expect_equal(class(b$mds_plot)[2], "ggplot")
39 
-  expect_equal(class(c$mds_plot)[2], "ggplot")
40 
-  expect_equal(class(d$mds_plot)[2], "ggplot")
41 
-  expect_equal(class(e$mds_plot)[2], "ggplot")
42 
-  expect_equal(class(f$mds_plot)[2], "ggplot")
43 
-
44 
-  ## optimum clusters
45 
-
46 
-  expect_equal(class(g$optimum_cluster_plot)[2], "ggplot")
47 
-  expect_equal(class(h$optimum_cluster_plot)[2], "ggplot")
48 
-
49 
-  # expect_equal(g$optimum_cluster_num, h$optimum_cluster_num)
50 
-
51 
-  expect_equal(5, ncol(g$mds_values))
52 
-  expect_equal(5, ncol(h$mds_values))
53 
-
54 
-  expect_false(length(levels(g$mds_values$clust)) == length(levels(h$mds_values$clust)))
55 
-
56 
-  ## errors
57 
-
58 
-  expect_error(PomaClust())
59 
-  expect_error(PomaClust(iris))
60 
-  expect_error(PomaClust(st000284, method =  "euclid"))
61 
-  expect_error(PomaClust(st000284, method =  "max"))
62 
-
2 
+test_that("PomaClust handles valid SummarizedExperiment objects", {
3 
+  data <- create_mock_summarized_experiment()
4 
+  result <- PomaClust(data)
5 
+  expect_is(result, "list")
6 
+  expect_true(all(c("mds_coordinates", "mds_plot", "optimal_clusters_number", "optimal_clusters_plot") %in% names(result)))
7 
+})
8 
+
9 
+test_that("PomaClust stops with non-SummarizedExperiment objects", {
10 
+  data <- data.frame(matrix(runif(100), ncol = 10))
11 
+  expect_error(PomaClust(data), "data is not a SummarizedExperiment object")
12 
+})
13 
+
14 
+test_that("PomaClust handles different methods correctly", {
15 
+  data <- create_mock_summarized_experiment()
16 
+  for (method in c("euclidean", "maximum", "manhattan", "canberra", "minkowski")) {
17 
+    result <- PomaClust(data, method = method)
18 
+    expect_is(result, "list")
19 
+  }
20 
+})
21 
+
22 
+test_that("PomaClust stops with incorrect method argument", {
23 
+  data <- create_mock_summarized_experiment()
24 
+  expect_error(PomaClust(data, method = "invalid_method"), "Incorrect value for method argument")
25 
+})
26 
+
27 
+test_that("PomaClust handles show_clusters and labels parameters correctly", {
28 
+  data <- create_mock_summarized_experiment()
29 
+  result_with_clusters <- PomaClust(data, show_clusters = TRUE)
30 
+  result_with_labels <- PomaClust(data, labels = TRUE)
31 
+  expect_is(result_with_clusters, "list")
32 
+  expect_is(result_with_labels, "list")
63 33 
 })
64 34
tests/testthat/test-PomaCorr.R
History View file @ 1b23d90
... ... 
@@ -1,72 +1,33 @@
1 
-context("PomaCorr")
2 1
3 
-test_that("PomaCorr works", {
4 
-
5 
-  library(ggraph)
6 
-
7 
-  data("st000284")
8 
-  data("st000336")
9 
-  imp_st000336 <- PomaImpute(st000336, method = "knn")
10 
-
11 
-  a <- PomaCorr(st000284)
12 
-  b <- PomaCorr(imp_st000336)
13 
-
14 
-  c <- PomaCorr(st000284, corr_type = "glasso", coeff = 0.3, method = "spearman")
15 
-  d <- PomaCorr(imp_st000336 , corr_type = "glasso", coeff = 0.5)
16 
-
17 
-  e <- PomaCorr(st000284, corr_type = "glasso", coeff = 0.5)
18 
-  f <- PomaCorr(st000284, coeff = 0.5)
19 
-
20 
-  ## table
21 
-
22 
-  expect_equal(((113*113)-113)/2, nrow(a$correlations))
23 
-  expect_equal(((30*30)-30)/2, nrow(b$correlations))
24 
-  expect_equal(((113*113)-113)/2, nrow(c$correlations))
25 
-  expect_equal(((30*30)-30)/2, nrow(d$correlations))
26 
-
27 
-  expect_equal(5, ncol(a$correlations))
28 
-  expect_equal(5, ncol(b$correlations))
29 
-  expect_equal(5, ncol(c$correlations))
30 
-  expect_equal(5, ncol(d$correlations))
31 
-
32 
-  expect_equal(class(e$data_glasso)[1], "tbl_df")
33 
-  expect_equal(ncol(e$data_glasso), 3)
34 
-  expect_equal(ncol(e$data_glasso) + 2, ncol(a$correlations))
35 
-  expect_equal(class(f$data_glasso), "NULL")
36 
-
37 
-  ## corrplot
38 
-
39 
-  expect_equal(class(a$corrplot)[2], "ggplot")
40 
-  expect_equal(class(b$corrplot)[2], "ggplot")
41 
-  expect_equal(class(c$corrplot)[2], "ggplot")
42 
-  expect_equal(class(d$corrplot)[2], "ggplot")
43 
-
44 
-  ## networks
45 
-
46 
-  expect_true(class(a$graph)[1] == "ggraph")
47 
-  expect_true(class(b$graph)[1] == "ggraph")
48 
-  expect_true(class(c$graph)[1] == "ggraph")
49 
-  expect_true(class(d$graph)[1] == "ggraph")
50 
-
51 
-  expect_true(113 > nrow(a$graph$data))
52 
-  expect_equal(6, ncol(a$graph$data))
53 
-
54 
-  expect_true(30 > nrow(b$graph$data))
55 
-  expect_equal(6, ncol(b$graph$data))
56 
-
57 
-  expect_true(113 > nrow(c$graph$data))
58 
-  expect_true(30 > nrow(d$graph$data))
59 
-
60 
-  ## errors
61 
-
62 
-  expect_error(PomaCorr())
63 
-  expect_error(PomaCorr(iris))
64 
-  expect_error(PomaCorr(st000284, corr_type = "co"))
65 
-  expect_error(PomaCorr(st000284, coeff = 2))
66 
-  expect_error(PomaCorr(st000284, coeff = -0.2))
67 
-  expect_error(PomaCorr(st000284, method = "pear"))
68 
-  expect_error(PomaCorr(st000284, corr_type = "cor", coeff = 1))
69 
-  expect_error(PomaCorr(st000284, corr_type = "glasso", coeff = 1))
70 
-
2 
+test_that("PomaCorr handles valid SummarizedExperiment objects", {
3 
+  data <- create_mock_summarized_experiment()
4 
+  result <- PomaCorr(data)
5 
+  expect_is(result, "list")
6 
+  expect_true(all(c("correlations", "corrplot") %in% names(result)))
7 
+})
8 
+
9 
+test_that("PomaCorr stops with non-SummarizedExperiment objects", {
10 
+  data <- data.frame(matrix(runif(100), ncol = 10))
11 
+  expect_error(PomaCorr(data), "data is not a SummarizedExperiment object")
12 
+})
13 
+
14 
+test_that("PomaCorr handles different methods correctly", {
15 
+  data <- create_mock_summarized_experiment()
16 
+  for (method in c("pearson", "kendall", "spearman")) {
17 
+    result <- PomaCorr(data, method = method)
18 
+    expect_is(result, "list")
19 
+    expect_true(all(c("correlations", "corrplot") %in% names(result)))
20 
+  }
21 
+})
22 
+
23 
+test_that("PomaCorr stops with incorrect method argument", {
24 
+  data <- create_mock_summarized_experiment()
25 
+  expect_error(PomaCorr(data, method = "invalid_method"), "Incorrect value for method argument")
26 
+})
27 
+
28 
+test_that("PomaCorr applies label_size and theme_params correctly", {
29 
+  data <- create_mock_summarized_experiment()
30 
+  result <- PomaCorr(data, label_size = 10, theme_params = list(base_size = 10))
31 
+  expect_is(result, "list")
71 32 
 })
72 33
tests/testthat/test-PomaCreateObject.R
History View file @ 1b23d90
... ... 
@@ -1,54 +1,22 @@
1 
-context("PomaCreateObject")
2 1
3 
-test_that("PomaCreateObject works", {
4 
-
5 
-  target <- data.frame(ID = c("One", "Two", "Three", "Four"),
6 
-                       Group = c("Trtd", "Ctrl", "Trtd", "Ctrl"),
7 
-                       Smoking = c(1,0,0,1))
8 
-
9 
-  target2 <- data.frame(ID = c("Five", "One", "Three", "Two"),
10 
-                        Group = c("Ctrl", "Trtd", "Trtd", "Ctrl"),
11 
-                        Smoking = c(0,0,0,1))
12 
-
13 
-  target_error <- as.matrix(target)
14 
-
15 
-  target_error_2 <- data.frame(ID = c("Five", "One", "Three", "Two"),
16 
-                               Group = c("Ctrl", "Trtd", "Trtd", "Ctrl"),
17 
-                               Smoking = c(0,0,NA,1))
18 
-
19 
-  features <- data.frame(Feat.1 = c(1,2,3,4), Feat.2 = c(6,3,7,3), Feat.3 = c(3,5,23,24))
20 
-  features_error <- data.frame(Feat.1 = c(1,2,3,4,5), Feat.2 = c(6,3,7,4,3), Feat.3 = c(3,4,5,23,24))
21 
-
22 
-  a <- PomaCreateObject(target, features)
23 
-  b <- PomaCreateObject(target2, features)
24 
-
25 
-  ##
26 
-
27 
-  expect_true(validObject(a))
28 
-  expect_true(validObject(b))
29 
-
30 
-  ##
31 
-
32 
-  expect_false(all(rownames(SummarizedExperiment::colData(a)) == rownames(SummarizedExperiment::colData(b))))
33 
-
34 
-  ##
2 
+test_that("PomaCreateObject handles features and metadata correctly", {
3 
+  mock_data <- create_mock_data()
4 
+  se_object <- PomaCreateObject(metadata = mock_data$metadata, features = mock_data$features)
5 
+  expect_is(se_object, "SummarizedExperiment")
6 
+})
35 7
36 
-  expect_error(PomaCreateObject(target_error, features))
37 
-  expect_error(PomaCreateObject(target, features_error))
38 
-  expect_error(PomaCreateObject(target))
39 
-  # expect_error(PomaCreateObject(features))
40 
-
41 
-  ##
42 
-
43 
-  expect_false(all(colnames(features) == rownames(SummarizedExperiment::assay(a))))
44 
-  expect_false(all(colnames(target)[2:3] == names(SummarizedExperiment::colData(a))))
45 
-  expect_false(all(colnames(target2)[2:3] == names(SummarizedExperiment::colData(b))))
8 
+test_that("PomaCreateObject handles missing metadata", {
9 
+  mock_data <- create_mock_data()
10 
+  se_object <- PomaCreateObject(features = mock_data$features)
11 
+  expect_is(se_object, "SummarizedExperiment")
12 
+})
46 13
47 
-  expect_true(ncol(target2) != ncol(SummarizedExperiment::colData(a)))
48 
-
49 
-  ##
50 
-
51 
-  expect_error(PomaCreateObject(target_error_2, features))
52 
-
14 
+test_that("PomaCreateObject handles factor_levels parameter correctly", {
15 
+  mock_data <- create_mock_data()
16 
+  mock_data$metadata$numeric_var <- 1:10
17 
+  se_object <- PomaCreateObject(metadata = mock_data$metadata, features = mock_data$features, factor_levels = 5)
18 
+  expect_is(se_object, "SummarizedExperiment")
19 
+  expect_true("numeric_var" %in% colnames(SummarizedExperiment::colData(se_object)))
20 
+  expect_true(is.numeric(SummarizedExperiment::colData(se_object)$numeric_var))
53 21 
 })
54 22
tests/testthat/test-PomaDESeq.R
History View file @ 1b23d90
55 23 
new file mode 100644
... ... 
@@ -0,0 +1,26 @@
1 
+
2 
+test_that("PomaDESeq handles valid SummarizedExperiment objects", {
3 
+  data <- create_mock_summarized_experiment(integers = TRUE)
4 
+  DESeq_results <- PomaDESeq(data)
5 
+  expect_is(DESeq_results, "tbl_df")
6 
+})
7 
+
8 
+test_that("PomaDESeq stops with non-SummarizedExperiment objects", {
9 
+  data <- data.frame(matrix(runif(100), ncol = 10))
10 
+  expect_error(PomaDESeq(data), "data is not a SummarizedExperiment object")
11 
+})
12 
+
13 
+test_that("PomaDESeq handles different adjust methods correctly", {
14 
+  data <- create_mock_summarized_experiment(integers = TRUE)
15 
+  for (adjust_method in c("fdr", "holm", "hochberg", "hommel", "bonferroni", "BH", "BY")) {
16 
+    DESeq_results <- PomaDESeq(data, adjust = adjust_method)
17 
+    expect_is(DESeq_results, "tbl_df")
18 
+  }
19 
+})
20 
+
21 
+test_that("PomaDESeq requires metadata", {
22 
+  data <- create_mock_summarized_experiment(integers = TRUE)
23 
+  metadata_removed_data <- SummarizedExperiment::SummarizedExperiment(assays = SummarizedExperiment::assay(data))
24 
+  expect_error(PomaDESeq(metadata_removed_data), "metadata file required")
25 
+})
26 
+
tests/testthat/test-PomaDensity.R
History View file @ 1b23d90
... ... 
@@ -1,55 +1,36 @@
1 
-context("PomaDensity")
2 1
3 
-test_that("PomaDensity works", {
4 
-
5 
-  data("st000284")
6 
-
7 
-  norm_none <- PomaNorm(st000284, method = "none")
8 
-  norm_ls <- PomaNorm(st000284, method = "log_scaling")
9 
-
10 
-  a <- PomaDensity(norm_none)
11 
-  b <- PomaDensity(norm_ls)
12 
-  c <- PomaDensity(norm_none, x = "features")
13 
-  d <- PomaDensity(norm_ls, x = "features")
14 
-
15 
-  e <- PomaDensity(norm_none, x = "samples")
16 
-
17 
-  f <- PomaDensity(norm_ls, x = "features", feature_name = "methyl_succinate")
18 
-  g <- PomaDensity(norm_ls, x = "features", feature_name = c("methyl_succinate", "linoleic_acid"))
19 
-
20 
-  df_a <- ggplot2::layer_data(a)
21 
-  df_b <- ggplot2::layer_data(b)
22 
-  df_c <- ggplot2::layer_data(c)
23 
-  df_d <- ggplot2::layer_data(d)
24 
-  df_e <- ggplot2::layer_data(e)
25 
-  df_f <- ggplot2::layer_data(f)
26 
-  df_g <- ggplot2::layer_data(g)
27 
-
28 
-  ####
2 
+test_that("PomaDensity handles valid SummarizedExperiment objects", {
3 
+  data <- create_mock_summarized_experiment()
4 
+  plot_samples <- PomaDensity(data, x = "samples")
5 
+  plot_features <- PomaDensity(data, x = "features")
6 
+  expect_is(plot_samples, "ggplot")
7 
+  expect_is(plot_features, "ggplot")
8 
+})
29 9
30 
-  expect_false(min(df_a$y) != min(df_c$y))
31 
-  expect_false(min(df_b$y) != min(df_d$y))
32 
-  expect_false(all(df_a$y == df_b$y))
33 
-  expect_false(all(df_c$y == df_d$y))
10 
+test_that("PomaDensity stops with non-SummarizedExperiment objects", {
11 
+  data <- data.frame(matrix(runif(100), ncol = 10))
12 
+  expect_error(PomaDensity(data), "data is not a SummarizedExperiment object")
13 
+})
34 14
35 
-  expect_false(all(df_f$y == df_g$y))
36 
-  expect_false(all(df_f$y == df_d$y))
15 
+test_that("PomaDensity stops with incorrect x argument", {
16 
+  data <- create_mock_summarized_experiment()
17 
+  expect_error(PomaDensity(data, x = "invalid_option"), "Incorrect value for x argument")
18 
+})
37 19
38 
-  expect_equal(df_a, df_e)
20 
+test_that("PomaDensity handles feature_name parameter correctly", {
21 
+  data <- create_mock_summarized_experiment()
22 
+  plot_specific_feature <- PomaDensity(data, x = "features", feature_name = c("V1"))
23 
+  expect_is(plot_specific_feature, "ggplot")
24 
+})
39 25
40 
-  expect_error(PomaDensity(norm_ls, x = "samp"))
41 
-  expect_error(PomaDensity(norm_ls, x = "features", feature_name = "hello"))
26 
+test_that("PomaDensity stops with non-existing feature names", {
27 
+  data <- create_mock_summarized_experiment()
28 
+  expect_error(PomaDensity(data, x = "features", feature_name = c("non_existing_feature")), "Features not found")
29 
+})
42 30
43 
-  expect_error(PomaDensity(norm_ls, feature_name = "hello"))
44 
-  expect_error(PomaDensity(norm_ls, feature_name = "methyl_succinat"))
45 
-
46 
-  ##
47 
-
48 
-  expect_error(PomaDensity(x = "sample"))
49 
-  expect_error(PomaDensity(iris, x = "sample"))
50 
-
51 
-  expect_message(PomaDensity(norm_ls, x = "features", feature_name = c("methyl_succinate", "linoleic_aci")))
52 
-  expect_message(PomaDensity(norm_ls, feature_name = c("methyl_succinat", "linoleic_acid")))
53 
-
31 
+test_that("PomaDensity applies theme parameters correctly", {
32 
+  data <- create_mock_summarized_experiment()
33 
+  plot_with_theme <- PomaDensity(data, theme_params = list(legend_title = TRUE))
34 
+  expect_is(plot_with_theme, "ggplot")
54 35 
 })
55 36
tests/testthat/test-PomaHeatmap.R
History View file @ 1b23d90
... ... 
@@ -1,29 +1,30 @@
1 
-context("PomaHeatmap")
2 1
3 
-test_that("PomaHeatmap works", {
4 
-
5 
-  data("st000284")
6 
-  data("st000336")
7 
-
8 
-  a <- PomaHeatmap(st000284, sample_names = TRUE, feature_names = FALSE, show_legend = TRUE)
9 
-  b <- PomaHeatmap(st000284, sample_names = FALSE, feature_names = FALSE, show_legend = FALSE)
10 
-
11 
-  c <- PomaHeatmap(st000336, sample_names = TRUE, feature_names = TRUE, show_legend = TRUE)
12 
-  d <- PomaHeatmap(st000336, sample_names = FALSE, feature_names = TRUE, show_legend = FALSE)
13 
-
14 
-  ##
15 
-
16 
-  expect_equal(class(a), class(b))
17 
-  expect_equal(class(b), class(d))
18 
-
19 
-  expect_false(length(a@matrix) == length(d@matrix))
20 
-
21 
-  expect_equal(length(a@row_order), ncol(t(SummarizedExperiment::assay(st000284))))
22 
-  expect_equal(length(d@row_order), ncol(t(SummarizedExperiment::assay(st000336))))
23 
-
24 
-  ##
25 
-
26 
-  expect_error(PomaHeatmap())
27 
-  expect_error(PomaHeatmap(iris))
28 
-
29 
-})
30 2 
\ No newline at end of file
3 
+test_that("PomaHeatmap handles valid SummarizedExperiment objects", {
4 
+  data <- create_mock_summarized_experiment()
5 
+  heatmap_plot <- PomaHeatmap(data)
6 
+  expect_is(heatmap_plot, "Heatmap")
7 
+})
8 
+
9 
+test_that("PomaHeatmap stops with non-SummarizedExperiment objects", {
10 
+  data <- data.frame(matrix(runif(100), ncol = 10))
11 
+  expect_error(PomaHeatmap(data), "data is not a SummarizedExperiment object")
12 
+})
13 
+
14 
+test_that("PomaHeatmap handles covariates correctly", {
15 
+  data <- create_mock_summarized_experiment()
16 
+  heatmap_plot_with_covs <- PomaHeatmap(data, covs = c("group"))
17 
+  expect_is(heatmap_plot_with_covs, "Heatmap")
18 
+})
19 
+
20 
+test_that("PomaHeatmap handles sample_names and feature_names parameters correctly", {
21 
+  data <- create_mock_summarized_experiment()
22 
+  heatmap_plot_with_names <- PomaHeatmap(data, sample_names = FALSE, feature_names = TRUE)
23 
+  expect_is(heatmap_plot_with_names, "Heatmap")
24 
+})
25 
+
26 
+test_that("PomaHeatmap handles show_legend parameter correctly", {
27 
+  data <- create_mock_summarized_experiment()
28 
+  heatmap_plot_no_legend <- PomaHeatmap(data, show_legend = FALSE)
29 
+  expect_is(heatmap_plot_no_legend, "Heatmap")
30 
+})
31 
+
tests/testthat/test-PomaImpute.R
History View file @ 1b23d90
... ... 
@@ -1,170 +1,37 @@
1 
-context("PomaImpute")
2 1
3 
-test_that("PomaImpute works", {
4 
-
5 
-  data("st000284")
6 
-
7 
-  data <- t(SummarizedExperiment::assay(st000284))
8 
-
9 
-  data <- data*round(runif(n = 1, min = 0.01, max = 0.99), 3) # just to create decimals
10 
-  data[1:4, 5] <- 0 # create some zeros in the first group
11 
-  data[73:77, 5] <- 0 # create some zeros in the second group
12 
-
13 
-  data[10:14, 5] <- NA # create some NA in the first group
14 
-  data[78:81, 5] <- NA # create some NA in the second group
15 
-
16 
-  colnames(data) <- gsub("_", ":", colnames(data))
17 
-
18 
-  target <- SummarizedExperiment::colData(st000284) %>%
19 
-    as.data.frame() %>%
20 
-    tibble::rownames_to_column()
21 
-  testimput <- PomaCreateObject(features = data, metadata = target)
22 
-
23 
-  a <- ncol(t(SummarizedExperiment::assay(PomaImpute(testimput, method = "knn", zeros_as_na = FALSE, remove_na = FALSE, cutoff = 8))))
24 
-  b <- ncol(t(SummarizedExperiment::assay(PomaImpute(testimput, method = "knn", zeros_as_na = TRUE, remove_na = FALSE, cutoff = 8))))
25 
-  c <- ncol(t(SummarizedExperiment::assay(PomaImpute(testimput, method = "knn", zeros_as_na = FALSE, remove_na = TRUE, cutoff = 8))))
26 
-  d <- ncol(t(SummarizedExperiment::assay(PomaImpute(testimput, method = "knn", zeros_as_na = TRUE, remove_na = TRUE, cutoff = 8))))
27 
-
28 
-  e <- ncol(t(SummarizedExperiment::assay(PomaImpute(testimput, method = "knn", zeros_as_na = FALSE, remove_na = TRUE, cutoff = 20))))
29 
-  f <- ncol(t(SummarizedExperiment::assay(PomaImpute(testimput, method = "knn", zeros_as_na = FALSE, remove_na = TRUE, cutoff = 10))))
30 
-
31 
-  g <- PomaImpute(testimput, method = "half_min", zeros_as_na = FALSE, remove_na = TRUE, cutoff = 20)
32 
-  h <- PomaImpute(testimput, method = "knn", zeros_as_na = FALSE, remove_na = TRUE, cutoff = 20)
33 
-
34 
-  i <- PomaImpute(testimput, method = "half_min", zeros_as_na = FALSE, remove_na = FALSE, cutoff = 1)
35 
-  j <- PomaImpute(testimput, method = "mean", zeros_as_na = FALSE, remove_na = FALSE, cutoff = 1)
36 
-  k <- PomaImpute(testimput, method = "median", zeros_as_na = FALSE, remove_na = FALSE, cutoff = 1)
37 
-
38 
-  l <- PomaImpute(testimput, method = "knn", zeros_as_na = FALSE, remove_na = TRUE, cutoff = 20)
39 
-  m <- PomaImpute(testimput, method = "knn")
40 
-
41 
-  n <- PomaImpute(testimput, method = "none", remove_na = FALSE, cutoff = 2)
42 
-  o <- PomaImpute(testimput, method = "none", remove_na = FALSE, cutoff = 5)
43 
-  p <- PomaImpute(testimput, method = "none", cutoff = 20)
44 
-  q <- PomaImpute(testimput, method = "min", cutoff = 20)
45 
-
46 
-  data2 <- t(SummarizedExperiment::assay(testimput))
47 
-
48 
-  data2[1:4, 5] <- 1000
49 
-  data2[73:77, 5] <- 1000
50 
-
51 
-  testimput2 <- PomaCreateObject(features = data2, target = target)
52 
-
53 
-  r <- PomaImpute(testimput2, method = "half_min")
54 
-  s <- PomaImpute(testimput2, method = "median")
55 
-  t <- PomaImpute(testimput2, method = "mean")
56 
-  u <- PomaImpute(testimput2, method = "min")
57 
-  v <- PomaImpute(testimput2, method = "knn")
58 
-
59 
-  SummarizedExperiment::assay(testimput)[5, 175:190] <- NA
60 
-  h_1 <- PomaImpute(testimput, method = "knn", zeros_as_na = FALSE, remove_na = TRUE, cutoff = 1)
61 
-
62 
-  ##
63 
-
64 
-  expect_equal(testimput@NAMES[1], h@NAMES[1])
65 
-  expect_equal(length(testimput@NAMES), length(h@NAMES))
66 
-  expect_false(length(testimput@NAMES) == length(h_1@NAMES))
67 
-
68 
-  expect_equal(testimput@NAMES[1], n@NAMES[1])
69 
-  expect_equal(length(testimput@NAMES), length(n@NAMES))
70 
-
71 
-  ##
72 
-
73 
-  expect_equal(a, b)
74 
-  expect_equal(b, c)
75 
-  expect_equal(a, c)
76 
-  expect_equal(a, d)
77 
-  expect_equal(b, d)
78 
-  expect_equal(c, d)
79 
-
80 
-  expect_equal(d, e)
81 
-  expect_equal(c, e)
82 
-  expect_equal(e, f)
83 
-
84 
-  expect_false(all(SummarizedExperiment::assay(g) == SummarizedExperiment::assay(h)))
85 
-  expect_equal(dim(g), dim(h))
86 
-
87 
-  expect_equal(dim(i), dim(j))
88 
-  expect_equal(dim(j), dim(k))
89 
-
90 
-  expect_false(all(SummarizedExperiment::assay(i) == SummarizedExperiment::assay(j)))
91 
-  expect_false(all(SummarizedExperiment::assay(j) == SummarizedExperiment::assay(k)))
92 
-  expect_false(all(SummarizedExperiment::assay(k) == SummarizedExperiment::assay(i)))
93 
-
94 
-  expect_equal(SummarizedExperiment::assay(l), SummarizedExperiment::assay(m))
95 
-  expect_equal(SummarizedExperiment::assay(n), SummarizedExperiment::assay(o))
96 
-  expect_true(all(SummarizedExperiment::assay(p) == SummarizedExperiment::assay(q)))
97 
-
98 
-  expect_equal(dim(r), dim(s))
99 
-  expect_equal(dim(s), dim(t))
100 
-  expect_equal(dim(t), dim(u))
101 
-  expect_equal(dim(u), dim(v))
102 
-
103 
-  ####
104 
-
105 
-  expect_false(all(SummarizedExperiment::assay(r) == SummarizedExperiment::assay(s)))
106 
-  expect_false(all(SummarizedExperiment::assay(r) == SummarizedExperiment::assay(t)))
107 
-  expect_false(all(SummarizedExperiment::assay(r) == SummarizedExperiment::assay(u)))
108 
-  expect_false(all(SummarizedExperiment::assay(r) == SummarizedExperiment::assay(v)))
109 
-
110 
-  expect_false(all(SummarizedExperiment::assay(s) == SummarizedExperiment::assay(t)))
111 
-  expect_false(all(SummarizedExperiment::assay(s) == SummarizedExperiment::assay(u)))
112 
-  expect_false(all(SummarizedExperiment::assay(s) == SummarizedExperiment::assay(v)))
113 
-
114 
-  expect_false(all(SummarizedExperiment::assay(t) == SummarizedExperiment::assay(u)))
115 
-  expect_false(all(SummarizedExperiment::assay(t) == SummarizedExperiment::assay(v)))
116 
-
117 
-  expect_false(all(SummarizedExperiment::assay(u) == SummarizedExperiment::assay(v)))
2 
+test_that("PomaImpute handles valid SummarizedExperiment objects", {
3 
+  data <- create_mock_summarized_experiment()
4 
+  imputed_data <- PomaImpute(data, method = "mean")
5 
+  expect_is(imputed_data, "SummarizedExperiment")
6 
+})
118 7
119 
-  ####
8 
+test_that("PomaImpute stops with non-SummarizedExperiment objects", {
9 
+  data <- data.frame(matrix(runif(100), ncol = 10))
10 
+  expect_error(PomaImpute(data), "data is not a SummarizedExperiment object")
11 
+})
120 12
121 
-  expect_error(PomaImpute(testimput, method = "non"))
122 
-  expect_message(PomaImpute(testimput))
123 
-  expect_message(PomaImpute(testimput2))
13 
+test_that("PomaImpute handles zeros_as_na parameter correctly", {
14 
+  data <- create_mock_summarized_experiment()
15 
+  imputed_data <- PomaImpute(data, zeros_as_na = TRUE)
16 
+  expect_is(imputed_data, "SummarizedExperiment")
17 
+})
124 18
125 
-  ####
126 
-
127 
-  expect_message(PomaImpute(st000284, method = "knn"))
128 
-  expect_message(PomaImpute(st000284, method = "random_forest"))
129 
-
130 
-  ##
131 
-
132 
-  expect_error(PomaImpute(method = "knn"))
133 
-  expect_error(PomaImpute(iris, method = "knn"))
134 
-
19 
+test_that("PomaImpute handles remove_na and cutoff parameters correctly", {
20 
+  data <- create_mock_summarized_experiment()
21 
+  imputed_data <- PomaImpute(data, remove_na = TRUE, cutoff = 50)
22 
+  expect_is(imputed_data, "SummarizedExperiment")
135 23 
 })
136 24
137 
-##################################################################
138 
-##################################################################
25 
+test_that("PomaImpute handles different imputation methods correctly", {
26 
+  data <- create_mock_summarized_experiment()
27 
+  for (method in c("none", "half_min", "median", "mean", "min", "knn")) { # "random_forest"
28 
+    imputed_data <- PomaImpute(data, method = method)
29 
+    expect_is(imputed_data, "SummarizedExperiment")
30 
+  }
31 
+})
139 32
140 
-# rfImpute fails many times in virtual machines because it generates a
141 
-# huge proximity matrix that sometimes needs >2 cores to run
142 
-#
143 
-# test_that("PomaImpute works skip on Appveyor", {
144 
-#
145 
-#   skip_on_appveyor() # rfImpute needs more than 2 cores to run and Appveyor only have 2
146 
-#
147 
-#   data("st000336")
148 
-#
149 
-#   a_2 <- PomaImpute(st000336, method = "half_min")
150 
-#   b_2 <- PomaImpute(st000336, method = "median")
151 
-#   c_2 <- PomaImpute(st000336, method = "mean")
152 
-#   d_2 <- PomaImpute(st000336, method = "min")
153 
-#   e_2 <- PomaImpute(st000336, method = "knn")
154 
-#   f_2 <- PomaImpute(st000336, method = "rf")
155 
-#
156 
-#   ##
157 
-#
158 
-#   expect_false(all(SummarizedExperiment::assay(a_2) == SummarizedExperiment::assay(b_2)))
159 
-#   expect_false(all(SummarizedExperiment::assay(b_2) == SummarizedExperiment::assay(c_2)))
160 
-#   expect_false(all(SummarizedExperiment::assay(c_2) == SummarizedExperiment::assay(d_2)))
161 
-#   expect_false(all(SummarizedExperiment::assay(d_2) == SummarizedExperiment::assay(e_2)))
162 
-#   expect_false(all(SummarizedExperiment::assay(e_2) == SummarizedExperiment::assay(f_2)))
163 
-#   expect_false(all(SummarizedExperiment::assay(f_2) == SummarizedExperiment::assay(a_2)))
164 
-#
165 
-#   ##
166 
-#
167 
-#   expect_error(PomaImpute(st000284, method = "rf"))
168 
-#
169 
-# })
33 
+test_that("PomaImpute stops with incorrect method argument", {
34 
+  data <- create_mock_summarized_experiment()
35 
+  expect_error(PomaImpute(data, method = "invalid_method"), "Incorrect value for method argument")
36 
+})
170 37
tests/testthat/test-PomaLM.R
History View file @ 1b23d90
171 38 
new file mode 100644
... ... 
@@ -0,0 +1,37 @@
1 
+
2 
+test_that("PomaLM handles valid SummarizedExperiment objects", {
3 
+  data <- create_mock_summarized_experiment()
4 
+  data$NumericVariable <- 1:20
5 
+  lm_results <- PomaLM(data)
6 
+  expect_is(lm_results, "list")
7 
+  expect_true("lm_table" %in% names(lm_results))
8 
+  expect_true("regression_plot" %in% names(lm_results))
9 
+})
10 
+
11 
+test_that("PomaLM stops with non-SummarizedExperiment objects", {
12 
+  data <- data.frame(matrix(runif(100), ncol = 10))
13 
+  expect_error(PomaLM(data), "data is not a SummarizedExperiment object")
14 
+})
15 
+
16 
+test_that("PomaLM handles specific independent variables", {
17 
+  data <- create_mock_summarized_experiment()
18 
+  data$NumericVariable <- 1:20
19 
+  lm_results <- PomaLM(data, x = c("V1", "V2"))
20 
+  expect_is(lm_results, "list")
21 
+})
22 
+
23 
+test_that("PomaLM handles specific dependent variables", {
24 
+  data <- create_mock_summarized_experiment()
25 
+  data$NumericVariable <- 1:20
26 
+  lm_results <- PomaLM(data, y = "NumericVariable")
27 
+  expect_is(lm_results, "list")
28 
+})
29 
+
30 
+test_that("PomaLM provides expected output structure", {
31 
+  data <- create_mock_summarized_experiment()
32 
+  data$NumericVariable <- 1:20
33 
+  lm_results <- PomaLM(data)
34 
+  expect_is(lm_results$lm_table, "tbl_df")
35 
+  expect_true(all(c("feature", "estimate", "std_err", "statistic", "pvalue", "adj_pvalue") %in% names(lm_results$lm_table)))
36 
+})
37 
+
tests/testthat/test-PomaLMM.R
History View file @ 1b23d90
0 38 
new file mode 100644
... ... 
@@ -0,0 +1,40 @@
1 
+
2 
+test_that("PomaLMM handles valid SummarizedExperiment objects", {
3 
+  data <- create_mock_summarized_experiment()
4 
+  lmm_results <- PomaLMM(data)
5 
+  expect_is(lmm_results, "list")
6 
+  expect_true("variances" %in% names(lmm_results))
7 
+  expect_true("variances_plot" %in% names(lmm_results))
8 
+})
9 
+
10 
+test_that("PomaLMM stops with non-SummarizedExperiment objects", {
11 
+  data <- data.frame(matrix(runif(100), ncol = 10))
12 
+  expect_error(PomaLMM(data), "data is not a SummarizedExperiment object")
13 
+})
14 
+
15 
+test_that("PomaLMM handles specific independent variables", {
16 
+  data <- create_mock_summarized_experiment()
17 
+  data$NumericVariable <- 1:20
18 
+  data$FactorVariable <- factor(c(rep("s", 10), rep("d", 10)))
19 
+  lmm_results <- PomaLMM(data, x = c("NumericVariable", "FactorVariable"))
20 
+  expect_is(lmm_results, "list")
21 
+})
22 
+
23 
+test_that("PomaLMM handles specific dependent variables", {
24 
+  data <- create_mock_summarized_experiment()
25 
+  lmm_results <- PomaLMM(data, y = c("V1", "V2"))
26 
+  expect_is(lmm_results, "list")
27 
+})
28 
+
29 
+test_that("PomaLMM stops with incorrect adjustment method", {
30 
+  data <- create_mock_summarized_experiment()
31 
+  expect_error(PomaLMM(data, adjust = "invalid"), "Incorrect value for adjust argument")
32 
+})
33 
+
34 
+test_that("PomaLMM provides expected output structure", {
35 
+  data <- create_mock_summarized_experiment()
36 
+  lmm_results <- PomaLMM(data)
37 
+  expect_is(lmm_results$variances, "tbl_df")
38 
+  expect_true(all(c("feature", "Residual", "(Intercept)") %in% names(lmm_results$variances)))
39 
+})
40 
+
tests/testthat/test-PomaLasso.R
History View file @ 1b23d90
... ... 
@@ -1,55 +1,37 @@
1 
-context("PomaLasso")
2 1
3 
-test_that("PomaLasso works", {
4 
-
5 
-  data("st000336")
6 
-
7 
-  normalized <- st000336 %>%
8 
-    POMA::PomaImpute(method = "knn") %>%
9 
-    POMA::PomaNorm(method = "log_scaling")
10 
-
11 
-  normalized_test <- normalized
12 
-  normalized_test_less <- normalized
13 
-  SummarizedExperiment::colData(normalized_test)[,1] <- c(rep("C", 30), rep("G", 20), rep("P", 7))
14 
-  SummarizedExperiment::colData(normalized_test_less)[,1] <- "Control"
15 
-
16 
-  lasso_res <- PomaLasso(normalized, alpha = 1, ntest = NULL, nfolds = 3, labels = TRUE)
17 
-  ridge_res <- PomaLasso(normalized, alpha = 0, ntest = NULL, nfolds = 10)
18 
-  enet_res <- PomaLasso(normalized, alpha = 0.5, ntest = NULL, nfolds = 5, labels = TRUE)
19 
-  lasso_self_lambda <- PomaLasso(normalized, alpha = 1, ntest = NULL, nfolds = 10, lambda = seq(0.002, 20, length.out = 100))
20 
-
21 
-  ## TABLES
22 
-
23 
-  expect_false(nrow(lasso_res$coefficients) == nrow(ridge_res$coefficients))
24 
-  expect_false(nrow(lasso_res$coefficients) == nrow(enet_res$coefficients))
25 
-
26 
-  expect_equal(ncol(lasso_res$coefficients), ncol(ridge_res$coefficients))
27 
-  expect_equal(ncol(ridge_res$coefficients), ncol(enet_res$coefficients))
28 
-  expect_equal(ncol(lasso_res$coefficients), ncol(lasso_self_lambda$coefficients))
29 
-
30 
-  ## PLOTS
2 
+test_that("PomaLasso handles valid SummarizedExperiment objects", {
3 
+  data <- create_mock_summarized_experiment(binary = TRUE)
4 
+  lasso_results <- PomaLasso(data, alpha = 1)
5 
+  expect_is(lasso_results, "list")
6 
+  expect_true(all(c("coefficients", "coefficients_plot", "cv_plot", "model") %in% names(lasso_results)))
7 
+})
31 8
32 
-  df_a <- ggplot2::layer_data(lasso_res$coefficientPlot)
33 
-  df_b <- ggplot2::layer_data(ridge_res$coefficientPlot)
34 
-  df_e <- ggplot2::layer_data(enet_res$coefficientPlot)
9 
+test_that("PomaLasso stops with non-SummarizedExperiment objects", {
10 
+  data <- data.frame(matrix(runif(100), ncol = 10))
11 
+  expect_error(PomaLasso(data), "data is not a SummarizedExperiment object")
12 
+})
35 13
36 
-  df_c <- ggplot2::layer_data(lasso_res$cvLassoPlot)
37 
-  df_d <- ggplot2::layer_data(ridge_res$cvLassoPlot)
14 
+test_that("PomaLasso handles different alpha values correctly", {
15 
+  data <- create_mock_summarized_experiment(binary = TRUE)
16 
+  for (alpha in c(0, 0.5, 1)) { # Testing Ridge, Elasticnet, and Lasso
17 
+    lasso_results <- PomaLasso(data, alpha = alpha)
18 
+    expect_is(lasso_results, "list")
19 
+    expect_true(all(c("coefficients", "coefficients_plot", "cv_plot", "model") %in% names(lasso_results)))
20 
+  }
21 
+})
38 22
39 
-  expect_false(length(df_a$y) == length(df_b$y))
40 
-  expect_false(length(df_c$y) == length(df_d$y))
41 
-  expect_false(length(df_a$y) == length(df_e$y))
23 
+test_that("PomaLasso stops with incorrect alpha argument", {
24 
+  data <- create_mock_summarized_experiment(binary = TRUE)
25 
+  expect_error(PomaLasso(data, alpha = -1), "alpha must be a number between 0 and 1")
26 
+  expect_error(PomaLasso(data, alpha = 2), "alpha must be a number between 0 and 1")
27 
+})
42 28
43 
-  ## ERRORS
44 
-
45 
-  expect_error(PomaLasso(normalized, alpha = 2))
46 
-  expect_error(PomaLasso(normalized, alpha = -0.5))
47 
-  expect_error(PomaLasso(iris, alpha = 1))
48 
-  expect_error(PomaLasso(normalized_test, alpha = 1))
49 
-  expect_error(PomaLasso(normalized_test_less, alpha = 1))
50 
-  expect_error(PomaLasso())
51 
-  expect_error(PomaLasso(normalized, ntest = 60))
52 
-  expect_error(PomaLasso(normalized, ntest = 2))
53 
-
29 
+test_that("PomaLasso handles ntest and nfolds parameters correctly", {
30 
+  data <- create_mock_summarized_experiment(binary = TRUE)
31 
+  lasso_results_with_ntest <- PomaLasso(data, ntest = 10)
32 
+  lasso_results_with_nfolds <- PomaLasso(data, nfolds = 5)
33 
+  expect_is(lasso_results_with_ntest, "list")
34 
+  expect_true("confusion_matrix" %in% names(lasso_results_with_ntest))
35 
+  expect_is(lasso_results_with_nfolds, "list")
54 36 
 })
55 37
tests/testthat/test-PomaLimma.R
History View file @ 1b23d90
... ... 
@@ -1,46 +1,27 @@
1 
-context("PomaLimma")
2 1
3 
-test_that("PomaLimma works", {
4 
-
5 
-  data("st000284")
6 
-  data("st000336")
7 
-
8 
-  a <- PomaLimma(st000284, contrast = "CRC-Polyp", covariates = FALSE, adjust = "fdr")
9 
-  b <- PomaLimma(st000284, contrast = "CRC-Polyp", covariates = FALSE, adjust = "bonferroni")
10 
-  c <- PomaLimma(st000284, contrast = "CRC-Polyp", covariates = TRUE, adjust = "fdr")
11 
-  d <- PomaLimma(st000284, contrast = "CRC-Polyp", covariates = TRUE, adjust = "bonferroni")
12 
-
13 
-  e <- PomaLimma(st000336, contrast = "Controls-DMD", covariates = FALSE, adjust = "fdr")
14 
-  f <- PomaLimma(st000336, contrast = "Controls-DMD", covariates = TRUE, adjust = "fdr")
15 
-
16 
-  ####
17 
-
18 
-  expect_error(PomaLimma(st000284, covariates = FALSE, adjust = "fdr"))
19 
-  expect_error(PomaLimma(st000284, contrast = NULL))
20 
-
21 
-  ####
22 
-
23 
-  expect_equal(dim(a), dim(b))
24 
-  expect_equal(dim(b), dim(c))
25 
-  expect_equal(dim(c), dim(d))
26 
-
27 
-  expect_false(all(a == b))
28 
-  expect_false(all(a == c))
2 
+test_that("PomaLimma handles valid SummarizedExperiment objects", {
3 
+  data <- create_mock_summarized_experiment()
4 
+  limma_results <- PomaLimma(data, contrast = "A-B")
5 
+  expect_is(limma_results, "tbl_df")
6 
+})
29 7
30 
-  expect_false(all(b == d))
8 
+test_that("PomaLimma stops with non-SummarizedExperiment objects", {
9 
+  data <- data.frame(matrix(runif(100), ncol = 10))
10 
+  expect_error(PomaLimma(data, contrast = "A-B"), "data is not a SummarizedExperiment object")
11 
+})
31 12
32 
-  expect_equal(dim(e), dim(f))
33 
-
34 
-  ####
13 
+test_that("PomaLimma handles different contrasts correctly", {
14 
+  data <- create_mock_summarized_experiment()
15 
+  contrast <- levels(SummarizedExperiment::colData(data)[,1])
16 
+  limma_results <- PomaLimma(data, contrast = paste(contrast[1], contrast[2], sep = "-"))
17 
+  expect_is(limma_results, "tbl_df")
18 
+})
35 19
36 
-  SummarizedExperiment::colData(st000284) <- SummarizedExperiment::colData(st000284)[1]
37 
-  expect_error(PomaLimma(st000284, contrast = "CRC-Polyp", covariates = TRUE, adjust = "fdr"))
38 
-  expect_error(PomaLimma(st000284, contrast = "CRC-Polyp", covariates = TRUE, adjust = "fd"))
39 
-
40 
-  ##
41 
-
42 
-  expect_error(PomaLimma(contrast = "CRC-Polyp"))
43 
-  expect_error(PomaLimma(iris, contrast = "CRC-Polyp"))
44 
-
20 
+test_that("PomaLimma handles parameters correctly", {
21 
+  data <- create_mock_summarized_experiment()
22 
+  limma_results_with_covs <- PomaLimma(data, contrast = "A-B")
23 
+  limma_results_adjusted <- PomaLimma(data, contrast = "A-B", adjust = "holm")
24 
+  expect_is(limma_results_with_covs, "tbl_df")
25 
+  expect_is(limma_results_adjusted, "tbl_df")
45 26 
 })
46 27
tests/testthat/test-PomaMultivariate.R
History View file @ 1b23d90
47 28 
deleted file mode 100644
... ... 
@@ -1,112 +0,0 @@
1 
-context("PomaMultivariate")
2 
-
3 
-test_that("PomaMultivariate works", {
4 
-
5 
-  data("st000284")
6 
-
7 
-  #### PCA
8 
-
9 
-  multivariate_pca_1 <- PomaMultivariate(st000284, method = "pca", components = 4,
10 
-                                    center = FALSE, scale = FALSE, labels = TRUE)
11 
-
12 
-  multivariate_pca_2 <- PomaMultivariate(st000284, method = "pca", components = 5,
13 
-                                    center = TRUE, scale = TRUE)
14 
-
15 
-  ##
16 
-
17 
-  expect_equal(nrow(multivariate_pca_1$scores), nrow(multivariate_pca_2$scores))
18 
-  expect_false(ncol(multivariate_pca_1$scores) == ncol(multivariate_pca_2$scores))
19 
-  expect_equal(ncol(multivariate_pca_1$scores), 4)
20 
-  expect_equal(ncol(multivariate_pca_2$scores), 5)
21 
-
22 
-  expect_equal(ncol(multivariate_pca_1$eigenvalues), ncol(multivariate_pca_2$eigenvalues))
23 
-  expect_false(nrow(multivariate_pca_1$eigenvalues) == nrow(multivariate_pca_2$eigenvalues))
24 
-
25 
-  ##
26 
-
27 
-  df_a <- ggplot2::layer_data(multivariate_pca_1$screeplot)
28 
-  df_b <- ggplot2::layer_data(multivariate_pca_1$scoresplot)
29 
-
30 
-  df_c <- ggplot2::layer_data(multivariate_pca_2$screeplot)
31 
-  df_d <- ggplot2::layer_data(multivariate_pca_2$scoresplot)
32 
-
33 
-  expect_false(length(df_a$y) == length(df_c$y))
34 
-  expect_false(length(df_b$y) == length(df_d$y))
35 
-
36 
-  #### PLSDA
37 
-
38 
-  multivariate_plsda_1 <- PomaMultivariate(st000284, method = "plsda", components = 3,
39 
-                                           center = TRUE, scale = TRUE,
40 
-                                           validation = "Mfold", folds = 5, nrepeat = 10, labels = TRUE)
41 
-
42 
-  multivariate_plsda_2 <- PomaMultivariate(st000284, method = "plsda", components = 4,
43 
-                                           center = TRUE, scale = TRUE,
44 
-                                           validation = "loo", folds = 5, nrepeat = 1, vip = 1)
45 
-
46 
-  ##
47 
-
48 
-  expect_equal(ncol(multivariate_plsda_1$errors_plsda), ncol(multivariate_plsda_2$errors_plsda))
49 
-  expect_false(nrow(multivariate_plsda_1$errors_plsda) == nrow(multivariate_plsda_2$errors_plsda))
50 
-
51 
-  expect_false(ncol(multivariate_plsda_1$vip_plsda) == ncol(multivariate_plsda_2$vip_plsda))
52 
-  expect_equal(nrow(multivariate_plsda_1$vip_plsda), nrow(multivariate_plsda_2$vip_plsda))
53 
-
54 
-  expect_false(ncol(multivariate_plsda_1$scores) == ncol(multivariate_plsda_2$scores))
55 
-
56 
-  ##
57 
-
58 
-  df_a <- ggplot2::layer_data(multivariate_plsda_1$scoresplot)
59 
-  df_b <- ggplot2::layer_data(multivariate_plsda_1$errors_plsda_plot)
60 
-  df_c <- ggplot2::layer_data(multivariate_plsda_1$vip_plsda_plot)
61 
-
62 
-  df_d <- ggplot2::layer_data(multivariate_plsda_2$scoresplot)
63 
-  df_e <- ggplot2::layer_data(multivariate_plsda_2$errors_plsda_plot)
64 
-  df_f <- ggplot2::layer_data(multivariate_plsda_2$vip_plsda_plot)
65 
-
66 
-  expect_false(ncol(df_a) == ncol(df_d))
67 
-  expect_equal(ncol(df_b$y), ncol(df_e$y))
68 
-  expect_false(length(df_c$y) == length(df_f$y))
69 
-
70 
-  #### SPLSDA
71 
-
72 
-  multivariate_splsda_1 <- PomaMultivariate(st000284, method = "splsda", components = 3,
73 
-                                            center = TRUE, scale = TRUE,
74 
-                                            validation = "Mfold", folds = 5, nrepeat = 10,
75 
-                                            num_features = 10, labels = TRUE)
76 
-
77 
-  multivariate_splsda_2 <- PomaMultivariate(st000284, method = "splsda", components = 4,
78 
-                                            center = TRUE, scale = TRUE,
79 
-                                            validation = "Mfold", folds = 5, nrepeat = 10,
80 
-                                            num_features = 5)
81 
-
82 
-  ##
83 
-
84 
-  expect_false(nrow(multivariate_splsda_1$selected_variables) ==
85 
-                 nrow(multivariate_splsda_2$selected_variables))
86 
-  expect_false(nrow(multivariate_splsda_1$errors_splsda) ==
87 
-                 nrow(multivariate_splsda_2$errors_splsda))
88 
-
89 
-  expect_true(is.numeric(multivariate_splsda_1$ncomp))
90 
-  expect_true(is.numeric(multivariate_splsda_2$ncomp))
91 
-
92 
-  df_a <- ggplot2::layer_data(multivariate_splsda_1$errors_splsda_plot)
93 
-  df_b <- ggplot2::layer_data(multivariate_splsda_2$errors_splsda_plot)
94 
-  df_c <- ggplot2::layer_data(multivariate_splsda_1$scoresplot)
95 
-  df_d <- ggplot2::layer_data(multivariate_splsda_2$scoresplot)
96 
-
97 
-  expect_false(nrow(df_a) == nrow(df_b))
98 
-  expect_false(length(df_c$y) == length(df_d$y))
99 
-
100 
-  ## ERRORS AND WARNINGS
101 
-
102 
-  expect_error(PomaMultivariate(method = "splsda"))
103 
-  expect_error(PomaMultivariate(iris, method = "splsda"))
104 
-  expect_error(PomaMultivariate(st000284, method = "pc", components = 5))
105 
-  expect_error(PomaMultivariate(st000284))
106 
-  expect_error(PomaMultivariate(st000284, method = "plsda", validation = "Mfo"))
107 
-  expect_message(PomaMultivariate(st000284, method = "plsda"))
108 
-  expect_error(PomaMultivariate(st000284, method = "pca", load_length = 2.1))
109 
-  expect_error(PomaMultivariate(st000284, method = "pca", load_length = 0.9))
110 
-
111 
-})
112 
-
tests/testthat/test-PomaNorm.R
History View file @ 1b23d90
... ... 
@@ -1,53 +1,34 @@
1 
-context("PomaNorm")
2 1
3 
-test_that("PomaNorm works", {
4 
-
5 
-  data("st000284")
6 
-
7 
-  data <- t(SummarizedExperiment::assay(st000284))
8 
-
9 
-  data <- data*round(runif(n=1, min = 0.01, max = 0.99), 3) # just to create decimals
10 
-  data[1:4, 5] <- 0 # create some zeros in one group
11 
-  data[73:77, 5] <- 0 # create some zeros in the other group
12 
-
13 
-  data[10:14, 5] <- NA # create some NA in one group (5/66 = 7.6% of NA)
14 
-  data[78:81, 5] <- NA # create some NA in the other group (4/66 = 6.1% of NA)
15 
-
16 
-  data[,1] <- 0 # create column of only zeros
17 
-  data[,2] <- 100 # create feature with var = 0
18 
-
19 
-  target <- SummarizedExperiment::colData(st000284) %>%
20 
-    as.data.frame() %>%
21 
-    tibble::rownames_to_column()
22 
-
23 
-  testnorm <- POMA::PomaCreateObject(features = data, metadata = target)
24 
-
25 
-  newdata <- POMA::PomaImpute(testnorm, method = "knn", zeros_as_na = FALSE, remove_na = TRUE, cutoff = 2)
26 
-  newdata2 <- POMA::PomaNorm(newdata, method = "log_pareto")
27 
-
28 
-  ####
2 
+test_that("PomaNorm handles valid SummarizedExperiment objects", {
3 
+  data <- create_mock_summarized_experiment()
4 
+  normalized_data <- PomaNorm(data, method = "auto_scaling")
5 
+  expect_is(normalized_data, "SummarizedExperiment")
6 
+})
29 7
30 
-  a <- dim(PomaNorm(newdata, method = "auto_scaling"))
31 
-  b <- dim(PomaNorm(newdata, method = "level_scaling"))
32 
-  c <- dim(PomaNorm(newdata, method = "log_scaling"))
33 
-  d <- dim(PomaNorm(newdata, method = "log_transform"))
34 
-  e <- dim(PomaNorm(newdata, method = "vast_scaling"))
35 
-  f <- dim(PomaNorm(newdata, method = "log_pareto"))
36 
-  g <- dim(PomaNorm(newdata, method = "none"))
8 
+test_that("PomaNorm stops with non-SummarizedExperiment objects", {
9 
+  data <- data.frame(matrix(runif(100), ncol = 10))
10 
+  expect_error(PomaNorm(data), "data is not a SummarizedExperiment object")
11 
+})
37 12
38 
-  expect_equal(a, b)
39 
-  expect_equal(b, c)
40 
-  expect_equal(c, d)
41 
-  expect_equal(e, f)
42 
-  expect_equal(f, g)
13 
+test_that("PomaNorm handles sample normalization methods correctly", {
14 
+  data <- create_mock_summarized_experiment()
15 
+  normalized_data_sum <- PomaNorm(data, sample_norm = "sum")
16 
+  normalized_data_quantile <- PomaNorm(data, sample_norm = "quantile")
17 
+  expect_is(normalized_data_sum, "SummarizedExperiment")
18 
+  expect_is(normalized_data_quantile, "SummarizedExperiment")
19 
+})
43 20
44 
-  expect_error(PomaNorm(newdata, method = "log"))
45 
-  expect_message(PomaNorm(newdata))
21 
+test_that("PomaNorm handles different normalization methods correctly", {
22 
+  data <- create_mock_summarized_experiment()
23 
+  for (method in c("none", "auto_scaling", "level_scaling", "log_scaling", "log_transform",
24 
+                   "vast_scaling", "log_pareto", "min_max", "box_cox")) {
25 
+    normalized_data <- PomaNorm(data, method = method)
26 
+    expect_is(normalized_data, "SummarizedExperiment")
27 
+  }
28 
+})
46 29
47 
-  ##
48 
-
49 
-  expect_error(PomaNorm(method = "auto_scaling"))
50 
-  expect_error(PomaNorm(iris, method = "auto_scaling"))
51 
-
30 
+test_that("PomaNorm stops with incorrect normalization method argument", {
31 
+  data <- create_mock_summarized_experiment()
32 
+  expect_error(PomaNorm(data, method = "invalid_method"), "Incorrect value for method argument")
52 33 
 })
53 34
tests/testthat/test-PomaOddsRatio.R
History View file @ 1b23d90
... ... 
@@ -1,82 +1,34 @@
1 
-context("PomaOddsRatio")
2 1
3 
-test_that("PomaOddsRatio works", {
4 
-
5 
-  data("st000336")
6 
-
7 
-  imputed <- POMA::PomaImpute(st000336, method = "knn")
8 
-
9 
-  norm_none <- PomaNorm(imputed, method = "none")
10 
-  norm_ls <- PomaNorm(imputed, method = "log_scaling")
11 
-
12 
-  a <- PomaOddsRatio(norm_none, feature_name = c("glutamic_acid", "glutamine", "glycine", "histidine", "isoleucine", "leucine", "lysine"))$OddsRatioPlot
13 
-  b <- PomaOddsRatio(norm_ls, feature_name = c("glutamic_acid", "glutamine", "glycine", "histidine", "isoleucine", "leucine", "lysine"))$OddsRatioPlot
14 
-
15 
-  c <- PomaOddsRatio(norm_ls, feature_name = "glutamic_acid")$OddsRatioPlot
16 
-  d <- PomaOddsRatio(norm_ls, feature_name = c("glutamic_acid", "arginine"))$OddsRatioPlot
17 
-
18 
-  df_a <- ggplot2::layer_data(a)
19 
-  df_b <- ggplot2::layer_data(b)
20 
-
21 
-  ##
22 
-
23 
-  expect_equal(nrow(df_a), nrow(df_b))
24 
-
25 
-  ##
26 
-
27 
-  e <- PomaOddsRatio(norm_none)$OddsRatioTable
28 
-  f <- PomaOddsRatio(norm_ls)$OddsRatioTable
29 
-  g <- PomaOddsRatio(norm_ls, feature_name = "glutamic_acid")$OddsRatioTable
30 
-  h <- PomaOddsRatio(norm_ls, feature_name = c("glutamic_acid", "arginine"))$OddsRatioTable
31 
-  h_1 <- PomaOddsRatio(norm_ls, feature_name = c("glutamic_acid", "arginine"), covariates = TRUE)$OddsRatioTable
32 
-
33 
-  ##
34 
-
35 
-  expect_equal(dim(e), dim(f))
36 
-  expect_false(nrow(f) == nrow(g))
37 
-  expect_false(nrow(g) == nrow(h))
38 
-  expect_false(nrow(h) == nrow(h_1))
39 
-
40 
-  ##
41 
-
42 
-  i <- PomaOddsRatio(norm_ls, feature_name = NULL, covariates = TRUE)$OddsRatioTable
43 
-  j <- PomaOddsRatio(norm_ls, feature_name = NULL, covariates = FALSE)$OddsRatioTable
44 
-
45 
-  ##
46 
-
47 
-  expect_false(nrow(i) == nrow(j))
48 
-
49 
-  ##
50 
-
51 
-  k <- PomaOddsRatio(norm_ls, feature_name = c("glutamic_acid", "glutamine", "glycine", "histidine", "isoleucine", "leucine", "lysine"))$OddsRatioPlot
52 
-  l <- PomaOddsRatio(norm_ls, feature_name = c("glutamic_acid", "glutamine", "glycine", "histidine", "isoleucine", "leucine", "lysine"), showCI = FALSE)$OddsRatioPlot
53 
-  m <- PomaOddsRatio(norm_ls, feature_name = c("glutamic_acid", "glutamine", "glycine", "histidine", "isoleucine", "leucine", "lysine"), showCI = FALSE, covariates = TRUE)$OddsRatioPlot
54 
-
55 
-  df_k <- ggplot2::layer_data(k)
56 
-  df_l <- ggplot2::layer_data(l)
57 
-  df_m <- ggplot2::layer_data(m)
58 
-
59 
-  ##
60 
-
61 
-  expect_equal(nrow(df_k), nrow(df_l))
62 
-
63 
-  ##
2 
+test_that("PomaOddsRatio handles valid SummarizedExperiment objects", {
3 
+  data <- create_mock_summarized_experiment(binary = TRUE)
4 
+  odds_ratio_results <- PomaOddsRatio(data)
5 
+  expect_is(odds_ratio_results, "list")
6 
+  expect_true(all(c("odds_ratio_table", "odds_ratio_plot") %in% names(odds_ratio_results)))
7 
+})
64 8
65 
-  expect_error(PomaOddsRatio(norm_ls, feature_name = "hello"))
66 
-  expect_error(PomaOddsRatio(norm_ls, feature_name = "glutamic_aci"))
67 
-
68 
-  expect_warning(PomaOddsRatio(norm_ls, feature_name = c("glutamic_aci", "arginine")))
69 
-  expect_warning(PomaOddsRatio(norm_ls, feature_name = c("glutamic_aci", "arginine", "glutamine")))
9 
+test_that("PomaOddsRatio stops with non-SummarizedExperiment objects", {
10 
+  data <- data.frame(matrix(runif(100), ncol = 10))
11 
+  expect_error(PomaOddsRatio(data), "data is not a SummarizedExperiment object")
12 
+})
70 13
71 
-  ##
14 
+test_that("PomaOddsRatio handles specific feature names correctly", {
15 
+  data <- create_mock_summarized_experiment(binary = TRUE)
16 
+  feature_name <- rownames(SummarizedExperiment::assay(data))[1:2]
17 
+  odds_ratio_results <- PomaOddsRatio(data, feature_name = feature_name)
18 
+  expect_is(odds_ratio_results, "list")
19 
+  expect_true(all(feature_name %in% odds_ratio_results$odds_ratio_table$feature))
20 
+})
72 21
73 
-  SummarizedExperiment::colData(imputed) <- SummarizedExperiment::colData(imputed)[1]
74 
-  expect_error(PomaOddsRatio(imputed, covariates = TRUE))
75 
-
76 
-  ##
77 
-
78 
-  expect_error(PomaOddsRatio())
79 
-  expect_error(PomaOddsRatio(iris))
22 
+test_that("PomaOddsRatio stops with incorrect feature names", {
23 
+  data <- create_mock_summarized_experiment(binary = TRUE)
24 
+  expect_error(PomaOddsRatio(data, feature_name = "non_existing_feature"))
25 
+})
80 26
27 
+test_that("PomaOddsRatio handles covariates and show_ci parameters correctly", {
28 
+  data <- create_mock_summarized_experiment(binary = TRUE)
29 
+  odds_ratio_results_with_covs <- PomaOddsRatio(data, covs = c("Group"))
30 
+  odds_ratio_results_no_ci <- PomaOddsRatio(data, show_ci = FALSE)
31 
+  expect_is(odds_ratio_results_with_covs, "list")
32 
+  expect_is(odds_ratio_results_no_ci, "list")
81 33 
 })
82 34
tests/testthat/test-PomaPCA.R
History View file @ 1b23d90
83 35 
new file mode 100644
... ... 
@@ -0,0 +1,25 @@
1 
+
2 
+test_that("PomaPCA handles valid SummarizedExperiment objects", {
3 
+  data <- create_mock_summarized_experiment()
4 
+  pca_results <- PomaPCA(data)
5 
+  expect_is(pca_results, "list")
6 
+  expect_true(all(c("factors", "factors_plot", "eigenvalues", "eigenvalues_plot", "loadings", "loadings_plot", "biplot") %in% names(pca_results)))
7 
+})
8 
+
9 
+test_that("PomaPCA stops with non-SummarizedExperiment objects", {
10 
+  data <- data.frame(matrix(runif(100), ncol = 10))
11 
+  expect_error(PomaPCA(data), "data is not a SummarizedExperiment object")
12 
+})
13 
+
14 
+test_that("PomaPCA handles different parameter settings correctly", {
15 
+  data <- create_mock_summarized_experiment()
16 
+  pca_results_center_scale <- PomaPCA(data, center = TRUE, scale = TRUE)
17 
+  pca_results_ncomp <- PomaPCA(data, ncomp = 2)
18 
+  pca_results_labels <- PomaPCA(data, labels = TRUE)
19 
+  pca_results_ellipse <- PomaPCA(data, ellipse = TRUE)
20 
+  expect_is(pca_results_center_scale, "list")
21 
+  expect_is(pca_results_ncomp, "list")
22 
+  expect_is(pca_results_labels, "list")
23 
+  expect_is(pca_results_ellipse, "list")
24 
+})
25 
+
tests/testthat/test-PomaPCR.R
History View file @ 1b23d90
0 26 
new file mode 100644
... ... 
@@ -0,0 +1,35 @@
1 
+
2 
+test_that("PomaPCR handles valid SummarizedExperiment objects", {
3 
+  data <- create_mock_summarized_experiment()
4 
+  data$NumericVariable <- 1:20
5 
+  pcr_results <- PomaPCR(data, y = "NumericVariable")
6 
+  expect_is(pcr_results, "tbl_df")
7 
+  expect_true(all(c("component", "estimate", "std_err", "statistic", "pvalue", "adj_pvalue") %in% names(pcr_results)))
8 
+})
9 
+
10 
+test_that("PomaPCR stops with non-SummarizedExperiment objects", {
11 
+  data <- data.frame(matrix(runif(100), ncol = 10))
12 
+  expect_error(PomaPCR(data, y = "NumericVariable"), "data is not a SummarizedExperiment object")
13 
+})
14 
+
15 
+test_that("PomaPCR handles different number of components", {
16 
+  data <- create_mock_summarized_experiment()
17 
+  data$NumericVariable <- 1:20
18 
+  pcr_results_2comp <- PomaPCR(data, ncomp = 2, y = "NumericVariable")
19 
+  pcr_results_5comp <- PomaPCR(data, ncomp = 5, y = "NumericVariable")
20 
+  expect_is(pcr_results_2comp, "tbl_df")
21 
+  expect_is(pcr_results_5comp, "tbl_df")
22 
+})
23 
+
24 
+test_that("PomaPCR stops with incorrect y variable", {
25 
+  data <- create_mock_summarized_experiment()
26 
+  data$InvalidVariable <- rep("a", 20)
27 
+  expect_error(PomaPCR(data, y = "InvalidVariable"), "No numeric variables to be used as dependent variable in metadata file")
28 
+})
29 
+
30 
+test_that("PomaPCR stops with incorrect adjust argument", {
31 
+  data <- create_mock_summarized_experiment()
32 
+  data$NumericVariable <- 1:20
33 
+  expect_error(PomaPCR(data, y = "NumericVariable", adjust = "invalid"), "Incorrect value for adjust argument")
34 
+})
35 
+
tests/testthat/test-PomaPLS.R
History View file @ 1b23d90
0 36 
new file mode 100644
... ... 
@@ -0,0 +1,38 @@
1 
+
2 
+test_that("PomaPLS handles valid SummarizedExperiment objects", {
3 
+  data <- create_mock_summarized_experiment()
4 
+  data$numeric_var <- 1:20
5 
+  pls_results <- PomaPLS(data, method = "pls")
6 
+  expect_is(pls_results, "list")
7 
+  expect_true(all(c("factors", "factors_plot", "loadings", "loadings_plot") %in% names(pls_results)))
8 
+})
9 
+
10 
+test_that("PomaPLS stops with non-SummarizedExperiment objects", {
11 
+  data <- data.frame(matrix(runif(100), ncol = 10))
12 
+  expect_error(PomaPLS(data, method = "pls"), "data is not a SummarizedExperiment object")
13 
+})
14 
+
15 
+test_that("PomaPLS handles different PLS methods correctly", {
16 
+  data <- create_mock_summarized_experiment()
17 
+  plsda_results <- PomaPLS(data, method = "plsda", ncomp = 3)
18 
+  splsda_results <- PomaPLS(data, method = "splsda", ncomp = 3, num_features = 3)
19 
+  expect_is(plsda_results, "list")
20 
+  expect_is(splsda_results, "list")
21 
+})
22 
+
23 
+test_that("PomaPLS handles different parameter settings correctly", {
24 
+  data <- create_mock_summarized_experiment()
25 
+  data$numeric_var <- 1:20
26 
+  pls_results_ncomp <- PomaPLS(data, method = "pls", ncomp = 3)
27 
+  plsda_results_labels <- PomaPLS(data, method = "plsda", ncomp = 3, labels = TRUE)
28 
+  splsda_results_ellipse <- PomaPLS(data, method = "splsda", ncomp = 3, num_features = 3, ellipse = TRUE)
29 
+  expect_is(pls_results_ncomp, "list")
30 
+  expect_is(plsda_results_labels, "list")
31 
+  expect_is(splsda_results_ellipse, "list")
32 
+})
33 
+
34 
+test_that("PomaPLS stops with incorrect method argument", {
35 
+  data <- create_mock_summarized_experiment()
36 
+  expect_error(PomaPLS(data, method = "invalid_method"), "Incorrect value for method argument")
37 
+})
38 
+
tests/testthat/test-PomaRankProd.R
History View file @ 1b23d90
... ... 
@@ -1,42 +1,28 @@
1 
-context("PomaRankProd")
2 1
3 
-test_that("PomaRankProd works", {
4 
-
5 
-  data("st000284")
6 
-
7 
-  target <- SummarizedExperiment::colData(st000284)[1:100,] %>%
8 
-    as.data.frame() %>%
9 
-    tibble::rownames_to_column("ID")
10 
-  e <- SummarizedExperiment::assay(st000284)[,1:100]
11 
-
12 
-  data <- PomaSummarizedExperiment(target = target, features = t(e))
13 
-
14 
-  toy_data <- POMA::PomaNorm(data, method = "log_scaling")
15 
-  SummarizedExperiment::colData(toy_data)$groups <- c(rep("C", 25), rep("G", 25))
16 
-
17 
-  ##
18 
-
19 
-  RP_one <- PomaRankProd(data, logged = TRUE, logbase = 2)
20 
-  RP_two <- PomaRankProd(data, logged = TRUE, logbase = 10)
21 
-
22 
-  RP_five <- PomaRankProd(data, cutoff = 0.05, method = "pfp")
23 
-  RP_six <- PomaRankProd(data, cutoff = 0.05, method = "pval")
24 
-
25 
-  ##
26 
-
27 
-  expect_error(PomaRankProd())
28 
-  expect_error(PomaRankProd(data, method = "pfd"))
29 
-  expect_error(PomaRankProd(toy_data))
30 
-  expect_message(PomaRankProd(data))
31 
-  expect_error(PomaRankProd(iris))
2 
+test_that("PomaRankProd handles valid SummarizedExperiment objects", {
3 
+  data <- create_mock_summarized_experiment(binary = TRUE)
4 
+  rank_prod_results <- PomaRankProd(data)
5 
+  expect_is(rank_prod_results, "list")
6 
+  expect_true(all(c("up_regulated", "down_regulated", "up_regulated_plot", "down_regulated_plot") %in% names(rank_prod_results)))
7 
+})
32 8
33 
-  ##
9 
+test_that("PomaRankProd stops with non-SummarizedExperiment objects", {
10 
+  data <- data.frame(matrix(runif(100), ncol = 10))
11 
+  expect_error(PomaRankProd(data), "data is not a SummarizedExperiment object")
12 
+})
34 13
35 
-  expect_equal(dim(RP_one$upregulated), dim(RP_two$upregulated))
36 
-  expect_equal(dim(RP_one$downregulated), dim(RP_two$downregulated))
14 
+test_that("PomaRankProd handles different parameters correctly", {
15 
+  data <- create_mock_summarized_experiment(binary = TRUE)
16 
+  rank_prod_results_logged <- PomaRankProd(data, logged = TRUE)
17 
+  rank_prod_results_cutoff <- PomaRankProd(data, cutoff = 0.1)
18 
+  rank_prod_results_method <- PomaRankProd(data, method = "pval")
19 
+  expect_is(rank_prod_results_logged, "list")
20 
+  expect_is(rank_prod_results_cutoff, "list")
21 
+  expect_is(rank_prod_results_method, "list")
22 
+})
37 23
38 
-  expect_false(dim(RP_five$upregulated)[1] == dim(RP_six$upregulated)[1])
39 
-  expect_false(dim(RP_five$downregulated)[1] == dim(RP_six$downregulated)[1])
40 
-
24 
+test_that("PomaRankProd stops with incorrect method argument", {
25 
+  data <- create_mock_summarized_experiment(binary = TRUE)
26 
+  expect_error(PomaRankProd(data, method = "invalid_method"), "Incorrect value for method argument")
41 27 
 })
42 28
tests/testthat/test-PomaUMAP.R
History View file @ 1b23d90
43 29 
new file mode 100644
... ... 
@@ -0,0 +1,33 @@
1 
+
2 
+test_that("PomaUMAP handles valid SummarizedExperiment objects", {
3 
+  data <- create_mock_summarized_experiment()
4 
+  umap_results <- PomaUMAP(data)
5 
+  expect_is(umap_results, "list")
6 
+  expect_true("umap_embeddings" %in% names(umap_results))
7 
+  expect_true("umap_plot" %in% names(umap_results))
8 
+})
9 
+
10 
+test_that("PomaUMAP stops with non-SummarizedExperiment objects", {
11 
+  data <- data.frame(matrix(runif(100), ncol = 10))
12 
+  expect_error(PomaUMAP(data), "data is not a SummarizedExperiment object")
13 
+})
14 
+
15 
+test_that("PomaUMAP works with different n_neighbors values", {
16 
+  data <- create_mock_summarized_experiment()
17 
+  umap_results <- PomaUMAP(data, n_neighbors = 5)
18 
+  expect_is(umap_results, "list")
19 
+})
20 
+
21 
+test_that("PomaUMAP works with different n_components values", {
22 
+  data <- create_mock_summarized_experiment()
23 
+  umap_results <- PomaUMAP(data, n_components = 3)
24 
+  expect_is(umap_results, "list")
25 
+})
26 
+
27 
+test_that("PomaUMAP provides expected output structure", {
28 
+  data <- create_mock_summarized_experiment()
29 
+  umap_results <- PomaUMAP(data)
30 
+  expect_is(umap_results$umap_embeddings, "tbl_df")
31 
+  expect_true(all(c("sample", "UMAP1", "UMAP2", "clust", "member_prob") %in% names(umap_results$umap_embeddings)))
32 
+})
33 
+
tests/testthat/test-PomaUnivariate.R
History View file @ 1b23d90
... ... 
@@ -1,68 +1,48 @@
1 
-context("PomaUnivariate")
2 1
3 
-test_that("PomaUnivariate works", {
4 
-
5 
-  library(SummarizedExperiment)
6 
-
7 
-  data("st000284")
8 
-  data("st000336")
9 
-
10 
-  # st000284_sub <- st000284[,st000284@colData$factors %in% c("CRC", "Healthy")]
11 
-  st000336 <- POMA::PomaImpute(st000336, method = "knn")
12 
-
13 
-  dims_for_ttest_and_mann <- c(ncol(t(SummarizedExperiment::assay(st000336))), 9)
14 
-  dims_for_aov <- c(ncol(t(SummarizedExperiment::assay(st000284))),
15 
-                    length(levels(as.factor(SummarizedExperiment::colData(st000284)[,1]))) + 6)
16 
-  dims_for_krusk <- c(ncol(t(SummarizedExperiment::assay(st000284))),
17 
-                      length(levels(as.factor(SummarizedExperiment::colData(st000284)[,1]))) + 7)
18 
-
19 
-  univ_ttest <- PomaUnivariate(st000336, method = "ttest", adjust = "fdr")
20 
-  univ_aov <- PomaUnivariate(st000284, covariates = FALSE, method = "anova", adjust = "fdr")
21 
-  univ_aov_cov <- PomaUnivariate(st000284, covariates = TRUE, method = "anova", adjust = "fdr")
22 
-  univ_aov_cov2 <- PomaUnivariate(st000284, covariates = TRUE, method = "anova", adjust = "bonferroni")
2 
+test_that("PomaUnivariate handles valid SummarizedExperiment objects", {
3 
+  data <- create_mock_summarized_experiment(binary = TRUE)
4 
+  univariate_results <- PomaUnivariate(data, method = "ttest")
5 
+  expect_is(univariate_results, "list")
6 
+  expect_true("result" %in% names(univariate_results))
7 
+})
23 8
24 
-  univ_mann <- PomaUnivariate(st000336, method = "mann", adjust = "fdr")
25 
-  univ_kruskal <- PomaUnivariate(st000284, method = "kruskal", adjust = "fdr")
26 
-  univ_kruskal2 <- PomaUnivariate(st000284, method = "kruskal", adjust = "BY")
9 
+test_that("PomaUnivariate stops with non-SummarizedExperiment objects", {
10 
+  data <- data.frame(matrix(runif(100), ncol = 10))
11 
+  expect_error(PomaUnivariate(data), "data is not a SummarizedExperiment object")
12 
+})
27 13
28 
-  one_cov1 <- PomaUnivariate(st000336, covariates = FALSE, method = "anova", adjust = "fdr")
29 
-  one_cov2 <- PomaUnivariate(st000336, covariates = TRUE, method = "anova", adjust = "fdr")
30 
-
31 
-  ##
14 
+test_that("PomaUnivariate handles different methods correctly for 2 groups", {
15 
+  data <- create_mock_summarized_experiment(binary = TRUE)
16 
+  for (method in c("ttest", "mann")) {
17 
+    univariate_results <- PomaUnivariate(data, method = method)
18 
+    expect_is(univariate_results, "list")
19 
+    expect_true("result" %in% names(univariate_results))
20 
+  }
21 
+})
32 22
33 
-  expect_equal(dims_for_ttest_and_mann, dim(univ_ttest))
34 
-  expect_false(dims_for_aov[2] == dim(univ_aov_cov)[2])
35 
-  expect_false(dims_for_aov[2] == dim(univ_aov_cov2)[2])
36 
-  expect_equal(dims_for_aov[1], dim(univ_aov_cov)[1])
37 
-  expect_equal(dims_for_aov[1], dim(univ_aov_cov2)[1])
38 
-  expect_equal(dims_for_aov, dim(univ_aov))
39 
-  expect_false(all(univ_aov_cov == univ_aov_cov2))
23 
+test_that("PomaUnivariate handles different methods correctly for more than 2 groups", {
24 
+  data <- create_mock_summarized_experiment()
25 
+  for (method in c("anova", "kruskal")) {
26 
+    univariate_results <- PomaUnivariate(data, method = method)
27 
+    expect_is(univariate_results, "list")
28 
+    expect_true("result" %in% names(univariate_results))
29 
+  }
30 
+})
40 31
41 
-  ##
32 
+test_that("PomaUnivariate stops with incorrect method argument", {
33 
+  data <- create_mock_summarized_experiment()
34 
+  expect_error(PomaUnivariate(data, method = "invalid_method"), "Incorrect value for method argument")
35 
+})
42 36
43 
-  expect_equal(dims_for_ttest_and_mann, dim(univ_mann))
44 
-  expect_equal(dim(univ_kruskal), dim(univ_kruskal2))
45 
-  expect_false(all(univ_kruskal == univ_kruskal2))
46 
-  expect_equal(dims_for_krusk, dim(univ_kruskal))
47 
-  expect_equal(dims_for_krusk, dim(univ_kruskal2))
48 
-
49 
-  expect_false(dim(one_cov1)[2] == dim(one_cov2)[2])
50 
-  expect_equal(dim(one_cov1)[1], dim(one_cov2)[1])
51 
-
52 
-  ##
53 
-
54 
-  expect_error(PomaUnivariate(st000284, covariates = TRUE, method = "anov", adjust = "fdr"))
55 
-  expect_error(PomaUnivariate(st000284, covariates = TRUE, adjust = "fdr"))
56 
-
57 
-  expect_error(PomaUnivariate(st000336, method = "ttest", adjust = "fd"))
58 
-
59 
-  SummarizedExperiment::colData(st000284) <- SummarizedExperiment::colData(st000284)[1]
60 
-  expect_error(PomaUnivariate(st000284, method = "anova", covariates = TRUE, adjust = "fdr"))
61 
-
62 
-  ##
63 
-
64 
-  expect_error(PomaUnivariate(method = "ttest"))
65 
-  expect_error(PomaUnivariate(iris, method = "ttest"))
37 
+test_that("PomaUnivariate handles paired, var_equal, and adjust parameters correctly", {
38 
+  data_binary <- create_mock_summarized_experiment(binary = TRUE)
39 
+  data_binary_paired <- create_mock_summarized_experiment(binary = TRUE, paired = TRUE)
66 40
41 
+  univariate_results_paired <- PomaUnivariate(data_binary_paired, method = "ttest", paired = TRUE)
42 
+  univariate_results_var_equal <- PomaUnivariate(data_binary, method = "ttest", var_equal = TRUE)
43 
+  univariate_results_adjusted <- PomaUnivariate(data_binary, method = "ttest", adjust = "holm")
44 
+  expect_is(univariate_results_paired, "list")
45 
+  expect_is(univariate_results_var_equal, "list")
46 
+  expect_is(univariate_results_adjusted, "list")
67 47 
 })
68 48
tests/testthat/test-PomaVolcano.R
History View file @ 1b23d90
... ... 
@@ -12,7 +12,7 @@ test_that("PomaVolcano stops with non-SummarizedExperiment objects", {
12 12
13 13 
 test_that("PomaVolcano handles different methods correctly", {
14 14 
   data <- create_mock_summarized_experiment(binary = TRUE)
15 
-  for (method in c("ttest", "mann", "limma")) { # DESeq
15 
+  for (method in c("ttest", "mann", "limma")) { # "DESeq"
16 16 
     plot <- PomaVolcano(data, method = method)
17 17 
     expect_is(plot, "ggplot")
18 18 
   }
... ... 
@@ -20,7 +20,7 @@ test_that("PomaVolcano handles different methods correctly", {
20 20
21 21 
 test_that("PomaVolcano stops with incorrect method argument", {
22 22 
   data <- create_mock_summarized_experiment(binary = TRUE)
23 
-  expect_error(PomaVolcano(data, method = "invalid_method"), "Incorrect value for method argument")
23 
+  expect_error(PomaVolcano(data, method = "invalid_method"))
24 24 
 })
25 25
26 26 
 test_that("PomaVolcano handles p-value adjustments correctly", {
... ... 
@@ -43,7 +43,7 @@ test_that("PomaVolcano handles different pval_cutoff and log2fc_cutoff values",
43 43 
 })
44 44
45 45 
 test_that("PomaVolcano handles labels, paired, and var_equal parameters correctly", {
46 
-  data <- create_mock_summarized_experiment(binary = TRUE)
46 
+  data <- create_mock_summarized_experiment(binary = TRUE, paired = TRUE)
47 47 
   plot_with_labels <- PomaVolcano(data, labels = TRUE)
48 48 
   plot_paired <- PomaVolcano(data, paired = TRUE)
49 49 
   plot_var_equal <- PomaVolcano(data, var_equal = TRUE)
vignettes/POMA-demo.Rmd
History View file @ 1b23d90
50 50 
deleted file mode 100644
... ... 
@@ -1,288 +0,0 @@
1 
-title: "POMA Workflow"
2 
-author:
3 
-- name: Pol Castellano-Escuder, Ph.D.
4 
-  affiliation: Duke University
5 
-  email: polcaes@gmail.com
6 
-date: "`r BiocStyle::doc_date()`"
7 
-output:
8 
-    BiocStyle::html_document
9 
-vignette: >
10 
-  %\VignetteIndexEntry{POMA Workflow}
11 
-  %\VignetteEngine{knitr::rmarkdown}
12 
-  %\usepackage[utf8]{inputenc}
13 
-  %\VignetteEncoding{UTF-8}
14 
-bibliography: ["POMA.bib"]
15 
-biblio-style: apalike
16 
-link-citations: true
17 
-
18 
-**Compiled date**: `r Sys.Date()`
19 
-
20 
-**Last edited**: 2023-11-20
21 
-
22 
-**License**: `r packageDescription("POMA")[["License"]]`
23 
-
24 
-```{r, include = FALSE}
25 
-knitr::opts_chunk$set(
26 
-  collapse = TRUE,
27 
-  # fig.align = "center",
28 
-  comment = ">"
29 
-)
30 
-```
31 
-
32 
-# Installation
33 
-
34 
-To install the Bioconductor version of the POMA package, run the following code:
35 
-
36 
-```{r, eval = FALSE}
37 
-# install.packages("BiocManager")
38 
-BiocManager::install("POMA")
39 
-```
40 
-
41 
-# Load POMA
42 
-
43 
-```{r, warning = FALSE, message = FALSE}
44 
-library(POMA)
45 
-```
46 
-
47 
-Additionally, you can load other useful packages for this vignette:
48 
-
49 
-```{r, warning = FALSE, message = FALSE}
50 
-library(ggplot2)
51 
-library(plotly)
52 
-```
53 
-
54 
-# The POMA Workflow
55 
-
56 
-The `POMA` package functions are organized into three sequential, distinct blocks: Data Preparation, Pre-processing, and Statistical Analysis.
57 
-
58 
-## Data Preparation
59 
-
60 
-The `SummarizedExperiment` package from Bioconductor offers well-defined computational data structures for representing various types of omics experiment data [@SummarizedExperiment]. Utilizing these data structures can significantly improve data analysis. `POMA` leverages `SummarizedExperiment` objects, enhancing the reusability of existing methods for this class and contributing to more robust and reproducible workflows.
61 
-
62 
-The workflow begins with either loading or creating a `SummarizedExperiment` object. Typically, your data might be stored in separate matrices and/or data frames. The `PomaCreateObject` function simplifies this step by quickly building a SummarizedExperiment object for you.
63 
-
64 
-```{r, eval = FALSE}
65 
-# create an SummarizedExperiment object from two separated data frames
66 
-target <- readr::read_csv("your_target.csv")
67 
-features <- readr::read_csv("your_features.csv")
68 
-
69 
-data <- PomaCreateObject(metadata = target, features = features)
70 
-```
71 
-
72 
-Alternatively, if your data is already in a `SummarizedExperiment` object, you can proceed directly to the pre-processing step. This vignette uses sample data provided in `POMA`.
73 
-
74 
-```{r, warning = FALSE, message = FALSE}
75 
-# load example data
76 
-data("st000336")
77 
-```
78 
-
79 
-```{r, warning = FALSE, message = FALSE}
80 
-st000336
81 
-```
82 
-
83 
-### Brief Description of the Example Data
84 
-
85 
-This dataset comprises 57 samples, 31 metabolites, 1 covariate, and 2 experimental groups (Controls and DMD) from a targeted LC/MS study.
86 
-
87 
-_Duchenne Muscular Dystrophy (DMD) is an X-linked recessive form of muscular dystrophy that affects males via a mutation in the gene for the muscle protein, dystrophin. Progression of the disease results in severe muscle loss, ultimately leading to paralysis and death. Steroid therapy has been a commonly employed method for reducing the severity of symptoms. This study aims to quantify the urine levels of amino acids and organic acids in patients with DMD both with and without steroid treatment. Track the progression of DMD in patients who have provided multiple urine samples._
88 
-
89 
-This data was obtained from [Metabolomics Workbench](https://www.metabolomicsworkbench.org/data/DRCCMetadata.php?Mode=Study&DataMode=AllData&StudyID=ST000336&StudyType=MS&ResultType=1#DataTabs).
90 
-
91 
-## Pre Processing
92 
-
93 
-This stage of the workflow is pivotal, as the decisions made here fundamentally influence the final statistical results. This phase is methodically segmented into three steps: Missing Value Imputation, Normalization, and Outlier Detection.
94 
-
95 
-### Missing Value Imputation
96 
-
97 
-In metabolomics studies, it's not uncommon for certain features to be undetectable or unquantifiable in some samples, owing to a variety of biological and technical factors [@imputation]. To address this prevalent issue, `POMA` provides a suite of seven distinct imputation methods, each designed to effectively handle missing data. The choice of method can significantly impact the subsequent analysis, so it's crucial to select the one that aligns best with the specific characteristics and requirements of your dataset. To perform imputation, simply execute the following code:
98 
-
99 
-```{r}
100 
-imputed <- st000336 %>%
101 
-  PomaImpute(method = "knn", zeros_as_na = TRUE, remove_na = TRUE, cutoff = 20)
102 
-
103 
-imputed
104 
-```
105 
-
106 
-### Normalization
107 
-
108 
-Following missing value imputation, the next crucial step in the process is Data Normalization. In mass spectrometry (MS) data, various factors can introduce significant variability that profoundly impacts the final statistical outcomes. Normalization is thus an essential step, as it corrects for these variations, ensuring that the data can be compared more reliably across.
109 
-
110 
-```{r}
111 
-normalized <- imputed %>%
112 
-  PomaNorm(method = "log_pareto")
113 
-
114 
-normalized
115 
-```
116 
-
117 
-#### Normalization effect
118 
-
119 
-Sometimes, you will be interested in _how the normalization process affect your data_?
120 
-
121 
-To answer this question, **POMA** offers two exploratory functions, `PomaBoxplots` and `PomaDensity`, that can help to understand the normalization process.
122 
-
123 
-`PomaBoxplots` generates boxplots for all samples or features (depending on the group factor) of a `SummarizedExperiment` object. Here, we can compare objects before and after normalization step.
124 
-
125 
-```{r, message = FALSE}
126 
-PomaBoxplots(imputed, group = "samples",
127 
-             jitter = FALSE,
128 
-             legend_position = "none") +
129 
-  ggplot2::ggtitle("Not Normalized") # data before normalization
130 
-```
131 
-
132 
-```{r, message = FALSE}
133 
-PomaBoxplots(normalized,
134 
-             group = "samples",
135 
-             jitter = FALSE,
136 
-             legend_position = "none") +
137 
-  ggplot2::ggtitle("Normalized") # data after normalization
138 
-```
139 
-
140 
-On the other hand, `PomaDensity` shows the distribution of all features before and after the normalization process.
141 
-
142 
-```{r, message = FALSE}
143 
-PomaDensity(imputed,
144 
-            group = "features",
145 
-            legend_position = "none") +
146 
-  ggplot2::ggtitle("Not Normalized") # data before normalization
147 
-```
148 
-
149 
-```{r, message = FALSE}
150 
-PomaDensity(normalized,
151 
-            group = "features") +
152 
-  ggplot2::ggtitle("Normalized") # data after normalization
153 
-```
154 
-
155 
-### Outlier Detection
156 
-
157 
-Finally, the last step of this block is the Outlier Detection. Outlers are defined as observations that are not concordant with those of the vast majority of the remaining data points. These values can have an enormous influence on the resultant statistical analysis, being a dangerous ground for all required assumptions in the most commonly applied parametric tests in mass spectrometry as well as for all also required assumptions in many regression techniques and predictive modeling approaches. **POMA** allows the analysis of outliers as well as the possibility to remove them from the analysis using different modulable parameters.
158 
-
159 
-Analyze and remove outliers running the following two lines of code.
160 
-
161 
-```{r}
162 
-PomaOutliers(normalized, do = "analyze")$polygon_plot # to explore
163 
-pre_processed <- PomaOutliers(normalized, do = "clean") # to remove outliers
164 
-pre_processed
165 
-```
166 
-
167 
-## Statistical Analysis
168 
-
169 
-Once the data have been pre-processed, you can start with the statistical analysis step! **POMA** offers many different statistical methods and possible combinations to compute. However, in this vignette we will comment only some of the most used.
170 
-
171 
-### Univariate Analysis
172 
-
173 
-**POMA** allows you to perform all of the most used univariate statistical methods in MS by using only one function! `PomaUnivariate` wrap 4 different univariate methods (ttest, ANOVA and ANCOVA, Wilcoxon test and Kruskal-Wallis Rank Sum Test) that you can perform changing only the "method" argument.
174 
-
175 
-#### T-test
176 
-
177 
-```{r}
178 
-PomaUnivariate(pre_processed, method = "ttest")
179 
-```
180 
-
181 
-You can also compute a volcano plot using the T-test results. _Note that we're using the non-normalized object to avoid negative values in our data._
182 
-
183 
-```{r}
184 
-PomaVolcano(imputed, pval = "adjusted")
185 
-```
186 
-
187 
-#### Wilcoxon Test
188 
-
189 
-```{r, warning = FALSE}
190 
-PomaUnivariate(pre_processed, method = "mann")
191 
-```
192 
-
193 
-### Limma
194 
-
195 
-Other of the wide used statistical methods in many different omics, such as epigenomics or transcriptomics, is **limma** [@limma]. **POMA** provides an easy use implementation of _limma_ you only have to specify the desired contrast to compute.
196 
-
197 
-```{r}
198 
-PomaLimma(pre_processed, contrast = "Controls-DMD", adjust = "fdr")
199 
-```
200 
-
201 
-### Multivariate Analysis
202 
-
203 
-On the other hand, multivariate analysis implemented in **POMA** is quite similar to the univariate approaches. `PomaMultivariate` allows users to compute a PCA, PLS-DA or sPLS-DA by changing only the "method" parameter. This function is based on **mixOmics** package [@mixOmics].
204 
-
205 
-#### Principal Component Analysis
206 
-
207 
-```{r}
208 
-poma_pca <- PomaMultivariate(pre_processed, method = "pca")
209 
-```
210 
-
211 
-```{r}
212 
-poma_pca$scoresplot +
213 
-  ggplot2::ggtitle("Scores Plot")
214 
-```
215 
-
216 
-#### PLS-DA
217 
-
218 
-```{r, warning = FALSE, message = FALSE, results = 'hide'}
219 
-poma_plsda <- PomaMultivariate(pre_processed, method = "plsda")
220 
-```
221 
-
222 
-```{r}
223 
-poma_plsda$scoresplot +
224 
-  ggplot2::ggtitle("Scores Plot")
225 
-```
226 
-
227 
-```{r}
228 
-poma_plsda$errors_plsda_plot +
229 
-  ggplot2::ggtitle("Error Plot")
230 
-```
231 
-
232 
-### Correlation Analysis
233 
-
234 
-Often, correlation analysis is used to explore and discover relationships and patterns within our data. `PomaCorr` provides a flexible and easy way to do that providing a table with all pairwise coorelations in the data, a correlogram and a correlation graph.
235 
-
236 
-```{r}
237 
-poma_cor <- PomaCorr(pre_processed, label_size = 8, coeff = 0.6)
238 
-poma_cor$correlations
239 
-poma_cor$corrplot
240 
-poma_cor$graph
241 
-```
242 
-
243 
-Alternatively, if you switch the "corr_type" parameter to "glasso", this function will compute a **Gaussian Graphical Model** using the **glmnet** package [@glasso].
244 
-
245 
-```{r}
246 
-PomaCorr(pre_processed, corr_type = "glasso", coeff = 0.6)$graph
247 
-```
248 
-
249 
-### Lasso, Ridge and Elasticnet
250 
-
251 
-**POMA** also provides a function to perform a Lasso, Ridge and Elasticnet regression for binary outcomes in a very intuitive and easy way. `PomaLasso` is based on **glmnet** package [@glmnet]. This function allows you to create a test subset in your data, evaluate the prediction of your models and export the model computed (it could be useful to perform prediction models with MS data). If "ntest" parameter is set to NULL, `PomaLasso` will use all observations to create the model (useful for feature selection).
252 
-
253 
-```{r}
254 
-# alpha = 1 for Lasso
255 
-PomaLasso(pre_processed, alpha = 1, labels = TRUE)$coefficientPlot
256 
-```
257 
-
258 
-### Random Forest
259 
-
260 
-Finally, the random forest algorithm is also implemented in **POMA**. `PomaRandForest` uses the **randomForest** package [@randomForest] to facilitate the implementation of the algorithm and creates automatically both test and train sets to compute and evaluate the resultant models.
261 
-
262 
-```{r}
263 
-poma_rf <- PomaRandForest(pre_processed, ntest = 10, nvar = 10)
264 
-poma_rf$error_tree
265 
-```
266 
-
267 
-Resultant random forest model confusion matrix for **test** set:
268 
-
269 
-```{r}
270 
-poma_rf$confusionMatrix$table
271 
-```
272 
-
273 
-Gini index plot for the top 10 predictors:
274 
-
275 
-```{r}
276 
-poma_rf$MeanDecreaseGini_plot
277 
-```
278 
-
279 
-# Session Information
280 
-
281 
-```{r}
282 
-sessionInfo()
283 
-```
284 
-
285 
-# References
286 
-
vignettes/POMA-normalization.Rmd
History View file @ 1b23d90
... ... 
@@ -1,5 +1,5 @@
1 1 
 ---
2 
-title: "POMA Normalization Methods"
2 
+title: "Normalization Methods"
3 3 
 author:
4 4 
 - name: Pol Castellano-Escuder, Ph.D.
5 5 
   affiliation: Duke University
... ... 
@@ -8,7 +8,7 @@ date: "`r BiocStyle::doc_date()`"
8 8 
 output:
9 9 
     BiocStyle::html_document
10 10 
 vignette: >
11 
-  %\VignetteIndexEntry{POMA Normalization Methods}
11 
+  %\VignetteIndexEntry{Normalization Methods}
12 12 
   %\VignetteEngine{knitr::rmarkdown}
13 13 
   %\usepackage[utf8]{inputenc}
14 14 
   %\VignetteEncoding{UTF-8}
... ... 
@@ -19,14 +19,14 @@ link-citations: true
19 19
20 20 
 **Compiled date**: `r Sys.Date()`
21 21
22 
-**Last edited**: 2023-11-20
22 
+**Last edited**: 2023-12-06
23 23
24 24 
 **License**: `r packageDescription("POMA")[["License"]]`
25 25
26 26 
 ```{r, include = FALSE}
27 27 
 knitr::opts_chunk$set(
28 28 
   collapse = TRUE,
29 
-  # fig.align = "center",
29 
+  fig.align = "center",
30 30 
   comment = ">"
31 31 
 )
32 32 
 ```
... ... 
@@ -49,12 +49,11 @@ library(patchwork)
49 49
50 50 
 # Load Data and Imputation
51 51
52 
-Let's create a cleaned `SummarizedExperiment` object from the sample `st000336` data to explore the normalization effects.
52 
+Let's create a cleaned `SummarizedExperiment` object from the sample data `st000336` to explore the normalization effects.
53 53
54 54 
 ```{r, warning = FALSE, comment = NA}
55 
-# imputation using the default method KNN
56 55 
 example_data <- st000336 %>%
57 
-  PomaImpute()
56 
+  PomaImpute() # KNN imputation
58 57
59 58 
 example_data
60 59 
 ```
... ... 
@@ -75,7 +74,7 @@ log_pareto <- PomaNorm(example_data, method = "log_pareto")
75 74
76 75 
 ## Normalization effect on data dimensions
77 76
78 
-When we check for the dimension of the data after normalization we can see that ALL methods have the same effect on data dimension. `PomaNorm` **only** change the data dimension when the data have **features that only have zeros** and when the data have **features with 0 variance**. Only in these two cases `PomaNorm` will remove features of the data, changing the data dimensions.
77 
+When we check for the dimension of the data after normalization we can see that all methods have the same effect on data dimension. `PomaNorm` **only** modifies the data dimension when the dataset contains **only-zero features** or **zero-variance features**.
79 78
80 79 
 ```{r, warning = FALSE}
81 80 
 dim(SummarizedExperiment::assay(none))
... ... 
@@ -89,52 +88,45 @@ dim(SummarizedExperiment::assay(log_pareto))
89 88
90 89 
 ## Normalization effect on samples
91 90
92 
-Here we can evaluate the different normalization effects on samples [@normalization].
91 
+Here we can evaluate the normalization effects on samples [@normalization].
93 92
94 93 
 ```{r, message = FALSE, warning = FALSE}
95 94 
 a <- PomaBoxplots(none,
96 
-                  x = "samples",
97 
-                  jitter = FALSE) +
95 
+                  x = "samples") +
98 96 
   ggplot2::ggtitle("Not Normalized")
99 97
100 98 
 b <- PomaBoxplots(auto_scaling,
101 99 
                   x = "samples",
102 
-                  jitter = FALSE,
103 100 
                   legend_position = "none") +
104 101 
   ggplot2::ggtitle("Auto Scaling") +
105 102 
   ggplot2::theme(axis.text.x = ggplot2::element_blank())
106 103
107 104 
 c <- PomaBoxplots(level_scaling,
108 105 
                   x = "samples",
109 
-                  jitter = FALSE,
110 106 
                   legend_position = "none") +
111 107 
   ggplot2::ggtitle("Level Scaling") +
112 108 
   ggplot2::theme(axis.text.x = ggplot2::element_blank())
113 109
114 110 
 d <- PomaBoxplots(log_scaling,
115 111 
                   x = "samples",
116 
-                  jitter = FALSE,
117 112 
                   legend_position = "none") +
118 113 
   ggplot2::ggtitle("Log Scaling") +
119 114 
   ggplot2::theme(axis.text.x = ggplot2::element_blank())
120 115
121 116 
 e <- PomaBoxplots(log_transformation,
122 117 
                   x = "samples",
123 
-                  jitter = FALSE,
124 118 
                   legend_position = "none") +
125 119 
   ggplot2::ggtitle("Log Transformation") +
126 120 
   ggplot2::theme(axis.text.x = ggplot2::element_blank())
127 121
128 122 
 f <- PomaBoxplots(vast_scaling,
129 123 
                   x = "samples",
130 
-                  jitter = FALSE,
131 124 
                   legend_position = "none") +
132 125 
   ggplot2::ggtitle("Vast Scaling") +
133 126 
   ggplot2::theme(axis.text.x = ggplot2::element_blank())
134 127
135 128 
 g <- PomaBoxplots(log_pareto,
136 129 
                   x = "samples",
137 
-                  jitter = FALSE,
138 130 
                   legend_position = "none") +
139 131 
   ggplot2::ggtitle("Log Pareto") +
140 132 
   ggplot2::theme(axis.text.x = ggplot2::element_blank())
... ... 
@@ -145,7 +137,7 @@ a
145 137
146 138 
 ## Normalization effect on features
147 139
148 
-Here we can evaluate the different normalization effects on features.
140 
+Here we can evaluate the normalization effects on features.
149 141
150 142 
 ```{r, message = FALSE, warning = FALSE}
151 143 
 h <- PomaDensity(none,
vignettes/POMA-workflow.Rmd
History View file @ 1b23d90
152 144 
new file mode 100644
... ... 
@@ -0,0 +1,282 @@
1 
+---
2 
+title: "Get Started"
3 
+author:
4 
+- name: Pol Castellano-Escuder, Ph.D.
5 
+  affiliation: Duke University
6 
+  email: polcaes@gmail.com
7 
+date: "`r BiocStyle::doc_date()`"
8 
+output:
9 
+    BiocStyle::html_document
10 
+vignette: >
11 
+  %\VignetteIndexEntry{Get Started}
12 
+  %\VignetteEngine{knitr::rmarkdown}
13 
+  %\usepackage[utf8]{inputenc}
14 
+  %\VignetteEncoding{UTF-8}
15 
+bibliography: ["POMA.bib"]
16 
+biblio-style: apalike
17 
+link-citations: true
18 
+---
19 
+
20 
+**Compiled date**: `r Sys.Date()`
21 
+
22 
+**Last edited**: 2023-12-06
23 
+
24 
+**License**: `r packageDescription("POMA")[["License"]]`
25 
+
26 
+```{r, include = FALSE}
27 
+knitr::opts_chunk$set(
28 
+  collapse = TRUE,
29 
+  fig.align = "center",
30 
+  comment = ">"
31 
+)
32 
+```
33 
+
34 
+# Installation
35 
+
36 
+To install the Bioconductor version of the POMA package, run the following code:
37 
+
38 
+```{r, eval = FALSE}
39 
+# install.packages("BiocManager")
40 
+BiocManager::install("POMA")
41 
+```
42 
+
43 
+# Load POMA
44 
+
45 
+```{r, warning = FALSE, message = FALSE}
46 
+library(POMA)
47 
+```
48 
+
49 
+Additionally, you can load other useful packages for this vignette:
50 
+
51 
+```{r, warning = FALSE, message = FALSE}
52 
+library(ggplot2)
53 
+library(plotly)
54 
+```
55 
+
56 
+# The POMA Workflow
57 
+
58 
+The `POMA` package functions are organized into three sequential, distinct blocks: Data Preparation, Pre-processing, and Statistical Analysis.
59 
+
60 
+## Data Preparation
61 
+
62 
+The `SummarizedExperiment` package from Bioconductor offers well-defined computational data structures for representing various types of omics experiment data [@SummarizedExperiment]. Utilizing these data structures can significantly improve data analysis. `POMA` leverages `SummarizedExperiment` objects, enhancing the reusability of existing methods for this class and contributing to more robust and reproducible workflows.
63 
+
64 
+The workflow begins with either loading or creating a `SummarizedExperiment` object. Typically, your data might be stored in separate matrices and/or data frames. The `PomaCreateObject` function simplifies this step by quickly building a SummarizedExperiment object for you.
65 
+
66 
+```{r, eval = FALSE}
67 
+# create an SummarizedExperiment object from two separated data frames
68 
+target <- readr::read_csv("your_target.csv")
69 
+features <- readr::read_csv("your_features.csv")
70 
+
71 
+data <- PomaCreateObject(metadata = target, features = features)
72 
+```
73 
+
74 
+Alternatively, if your data is already in a `SummarizedExperiment` object, you can proceed directly to the pre-processing step. This vignette uses example data provided in `POMA`.
75 
+
76 
+```{r, warning = FALSE, message = FALSE}
77 
+# load example data
78 
+data("st000336")
79 
+```
80 
+
81 
+```{r, warning = FALSE, message = FALSE}
82 
+st000336
83 
+```
84 
+
85 
+<!-- ### Brief Description of the Example Data -->
86 
+
87 
+<!-- This dataset comprises 57 samples, 31 metabolites, 1 covariate, and 2 experimental groups (Controls and DMD) from a targeted LC/MS study.   -->
88 
+
89 
+<!-- _Duchenne Muscular Dystrophy (DMD) is an X-linked recessive form of muscular dystrophy that affects males via a mutation in the gene for the muscle protein, dystrophin. Progression of the disease results in severe muscle loss, ultimately leading to paralysis and death. Steroid therapy has been a commonly employed method for reducing the severity of symptoms. This study aims to quantify the urine levels of amino acids and organic acids in patients with DMD both with and without steroid treatment. Track the progression of DMD in patients who have provided multiple urine samples._     -->
90 
+
91 
+<!-- This data was obtained from [Metabolomics Workbench](https://www.metabolomicsworkbench.org/data/DRCCMetadata.php?Mode=Study&DataMode=AllData&StudyID=ST000336&StudyType=MS&ResultType=1#DataTabs).     -->
92 
+
93 
+## Pre Processing
94 
+
95 
+<!-- This stage of the workflow is pivotal, as the decisions made here fundamentally influence the final statistical results. This phase is methodically segmented into three steps: Missing Value Imputation, Normalization, and Outlier Detection. -->
96 
+
97 
+### Missing Value Imputation
98 
+
99 
+<!-- In metabolomics studies, it's not uncommon for certain features to be undetectable or unquantifiable in some samples, owing to a variety of biological and technical factors [@imputation]. To address this prevalent issue, `POMA` provides a suite of seven distinct imputation methods, each designed to effectively handle missing data. The choice of method can significantly impact the subsequent analysis, so it's crucial to select the one that aligns best with the specific characteristics and requirements of your dataset. To perform imputation, simply execute the following code: -->
100 
+
101 
+```{r}
102 
+imputed <- st000336 %>%
103 
+  PomaImpute(method = "knn", zeros_as_na = TRUE, remove_na = TRUE, cutoff = 20)
104 
+
105 
+imputed
106 
+```
107 
+
108 
+### Normalization
109 
+
110 
+<!-- Following missing value imputation, the next crucial step in the process is Data Normalization. In mass spectrometry (MS) data, various factors can introduce significant variability that profoundly impacts the final statistical outcomes. Normalization is thus an essential step, as it corrects for these variations, ensuring that the data can be compared more reliably across.       -->
111 
+
112 
+```{r}
113 
+normalized <- imputed %>%
114 
+  PomaNorm(method = "log_pareto")
115 
+
116 
+normalized
117 
+```
118 
+
119 
+#### Normalization effect
120 
+
121 
+<!-- `PomaBoxplots` generates boxplots for all samples or features of a `SummarizedExperiment` object. Here, we can compare objects before and after normalization step.     -->
122 
+
123 
+```{r, message = FALSE}
124 
+PomaBoxplots(imputed,
125 
+             x = "samples",
126 
+             legend_position = "none") +
127 
+  ggplot2::ggtitle("Not Normalized") # data before normalization
128 
+```
129 
+
130 
+```{r, message = FALSE}
131 
+PomaBoxplots(normalized,
132 
+             x = "samples",
133 
+             legend_position = "none") +
134 
+  ggplot2::ggtitle("Normalized") # data after normalization
135 
+```
136 
+
137 
+<!-- On the other hand, `PomaDensity` shows the distribution of all features before and after the normalization process.     -->
138 
+
139 
+```{r, message = FALSE}
140 
+PomaDensity(imputed,
141 
+            x = "features") +
142 
+  ggplot2::ggtitle("Not Normalized") # data before normalization
143 
+```
144 
+
145 
+```{r, message = FALSE}
146 
+PomaDensity(normalized,
147 
+            x = "features") +
148 
+  ggplot2::ggtitle("Normalized") # data after normalization
149 
+```
150 
+
151 
+### Outlier Detection
152 
+
153 
+<!-- Finally, the last step of this block is the Outlier Detection. Outlers are defined as observations that are not concordant with those of the vast majority of the remaining data points. These values can have an enormous influence on the resultant statistical analysis, being a dangerous ground for all required assumptions in the most commonly applied parametric tests in mass spectrometry as well as for all also required assumptions in many regression techniques and predictive modeling approaches. **POMA** allows the analysis of outliers as well as the possibility to remove them from the analysis using different modulable parameters.     -->
154 
+
155 
+<!-- Analyze and remove outliers running the following two lines of code.   -->
156 
+
157 
+```{r}
158 
+PomaOutliers(normalized)$polygon_plot
159 
+pre_processed <- PomaOutliers(normalized)$data
160 
+pre_processed
161 
+```
162 
+
163 
+## Statistical Analysis
164 
+
165 
+<!-- Once the data have been pre-processed, you can start with the statistical analysis step! **POMA** offers many different statistical methods and possible combinations to compute. However, in this vignette we will comment only some of the most used.     -->
166 
+
167 
+### Univariate Analysis
168 
+
169 
+<!-- `PomaUnivariate` computes four univariate methods (ttest, ANOVA and ANCOVA, Wilcoxon test and Kruskal-Wallis Rank Sum Test) that you can perform changing only the "method" argument.    -->
170 
+
171 
+#### T-test
172 
+
173 
+```{r}
174 
+PomaUnivariate(pre_processed, method = "ttest")
175 
+```
176 
+
177 
+<!-- You can also compute a volcano plot using the T-test results.    -->
178 
+
179 
+```{r}
180 
+PomaVolcano(imputed, pval = "adjusted", labels = TRUE)
181 
+```
182 
+
183 
+#### Wilcoxon Test
184 
+
185 
+```{r, warning = FALSE}
186 
+PomaUnivariate(pre_processed, method = "mann")
187 
+```
188 
+
189 
+<!-- ### Limma -->
190 
+
191 
+<!-- Other of the wide used statistical methods in many different omics, such as epigenomics or transcriptomics, is **limma** [@limma]. **POMA** provides an easy use implementation of _limma_ you only have to specify the desired contrast to compute.      -->
192 
+
193 
+```{r}
194 
+# PomaLimma(pre_processed, contrast = "Controls-DMD", adjust = "fdr")
195 
+```
196 
+
197 
+<!-- ### Multivariate Analysis -->
198 
+
199 
+<!-- On the other hand, multivariate analysis implemented in **POMA** is quite similar to the univariate approaches. `PomaMultivariate` allows users to compute a PCA, PLS-DA or sPLS-DA by changing only the "method" parameter. This function is based on **mixOmics** package [@mixOmics].     -->
200 
+
201 
+<!-- #### Principal Component Analysis -->
202 
+
203 
+```{r}
204 
+# poma_pca <- PomaMultivariate(pre_processed, method = "pca")
205 
+```
206 
+
207 
+```{r}
208 
+# poma_pca$scoresplot +
209 
+#   ggplot2::ggtitle("Scores Plot")
210 
+```
211 
+
212 
+<!-- #### PLS-DA -->
213 
+
214 
+```{r, warning = FALSE, message = FALSE, results = 'hide'}
215 
+# poma_plsda <- PomaMultivariate(pre_processed, method = "plsda")
216 
+```
217 
+
218 
+```{r}
219 
+# poma_plsda$scoresplot +
220 
+#   ggplot2::ggtitle("Scores Plot")
221 
+```
222 
+
223 
+```{r}
224 
+# poma_plsda$errors_plsda_plot +
225 
+#   ggplot2::ggtitle("Error Plot")
226 
+```
227 
+
228 
+<!-- ### Correlation Analysis -->
229 
+
230 
+<!-- Often, correlation analysis is used to explore and discover relationships and patterns within our data. `PomaCorr` provides a flexible and easy way to do that providing a table with all pairwise coorelations in the data, a correlogram and a correlation graph.     -->
231 
+
232 
+```{r}
233 
+# poma_cor <- PomaCorr(pre_processed, label_size = 8, coeff = 0.6)
234 
+# poma_cor$correlations
235 
+# poma_cor$corrplot
236 
+# poma_cor$graph
237 
+```
238 
+
239 
+<!-- Alternatively, if you switch the "corr_type" parameter to "glasso", this function will compute a **Gaussian Graphical Model** using the **glmnet** package [@glasso].    -->
240 
+
241 
+```{r}
242 
+# PomaCorr(pre_processed, corr_type = "glasso", coeff = 0.6)$graph
243 
+```
244 
+
245 
+<!-- ### Lasso, Ridge and Elasticnet -->
246 
+
247 
+<!-- **POMA** also provides a function to perform a Lasso, Ridge and Elasticnet regression for binary outcomes in a very intuitive and easy way. `PomaLasso` is based on **glmnet** package [@glmnet]. This function allows you to create a test subset in your data, evaluate the prediction of your models and export the model computed (it could be useful to perform prediction models with MS data). If "ntest" parameter is set to NULL, `PomaLasso` will use all observations to create the model (useful for feature selection).     -->
248 
+
249 
+```{r}
250 
+# alpha = 1 for Lasso
251 
+# PomaLasso(pre_processed, alpha = 1, labels = TRUE)$coefficientPlot
252 
+```
253 
+
254 
+<!-- ### Random Forest -->
255 
+
256 
+<!-- Finally, the random forest algorithm is also implemented in **POMA**. `PomaRandForest` uses the **randomForest** package [@randomForest] to facilitate the implementation of the algorithm and creates automatically both test and train sets to compute and evaluate the resultant models.      -->
257 
+
258 
+```{r}
259 
+# poma_rf <- PomaRandForest(pre_processed, ntest = 10, nvar = 10)
260 
+# poma_rf$error_tree
261 
+```
262 
+
263 
+<!-- Resultant random forest model confusion matrix for **test** set:   -->
264 
+
265 
+```{r}
266 
+# poma_rf$confusionMatrix$table
267 
+```
268 
+
269 
+<!-- Gini index plot for the top 10 predictors:   -->
270 
+
271 
+```{r}
272 
+# poma_rf$MeanDecreaseGini_plot
273 
+```
274 
+
275 
+# Session Information
276 
+
277 
+```{r}
278 
+sessionInfo()
279 
+```
280 
+
281 
+# References
282 
+