@@ -14,3 +14,4 @@ manuscript/

 .ignore

 R/immcantation-utils.R

 ^WORDLIST$

+^doc$

@@ -83,15 +83,16 @@ cdhit = function(seqs, identity = NULL, kmerSize = NULL, min_length = 6, s = 1,

     options = c(uopts, options)

     options = options[!duplicated(names(options))]

     options = lapply(options, as.character)

-    out = switch(

-        class(seqs),

-        AAStringSet = cdhitC(options, name, showProgress) + 1,

-        DNAStringSet = cdhitestC(options, name, showProgress) + 1,

-        stop('seqs must be either AAStringSet or DNAStringSet')

-    )

-    if(only_index) return(out)

-    tibble::tibble(query_name = names(seqs), seq = as.character(seqs), cluster_idx = out) %>%

-        dplyr::group_by(cluster_idx) %>% dplyr::mutate(n_cluster = dplyr::n())

+    if(type == 'cdhitC'){

+        seq_cluster_index = cdhitC(options, name, showProgress) + 1

+    } else{

+        seq_cluster_index = cdhitestC(options, name, showProgress) + 1

+    }

+    if(only_index) return(seq_cluster_index)

+    tibble::tibble(query_name = names(seqs), seq = as.character(seqs),

+                   cluster_idx = seq_cluster_index) %>%

+        dplyr::group_by(cluster_idx) %>%

+        dplyr::mutate(n_cluster = dplyr::n()) %>% ungroup()

 }

@@ -100,7 +101,7 @@ cdhit = function(seqs, identity = NULL, kmerSize = NULL, min_length = 6, s = 1,

 ##' @param ccdb An object of class [ContigCellDB()]

 ##' @param sequence_key `character` naming the column in the `contig_tbl` containing the sequence to be clustered

 ##' @param type one of 'DNA' or 'AA'

-##' @param cluster_name `character` specifying key, and name for the clustering.

+##' @param cluster_pk `character` specifying key, and name for the clustering.

 ##' @return [ContigCellDB()]

 ##' @inheritDotParams cdhit -seqs -only_index

 ##' @export

@@ -113,7 +114,7 @@ cdhit = function(seqs, identity = NULL, kmerSize = NULL, min_length = 6, s = 1,

 ##' res$contig_tbl

 ##' res$cluster_pk

 cdhit_ccdb = function(ccdb, sequence_key, type = c('DNA', 'AA'),

-                      cluster_name = 'cluster_idx', ...){

+                      cluster_pk = 'cluster_idx', ...){

     seqs = ccdb$contig_tbl[[sequence_key]]

     if(length(seqs) < 1) stop("No sequences were provided")

     type = match.arg(type, c('DNA', 'AA'))

@@ -123,7 +124,8 @@ cdhit_ccdb = function(ccdb, sequence_key, type = c('DNA', 'AA'),

         seqset = AAStringSet(seqs)

     }

     cluster_idx = cdhit(seqset, ..., only_index = TRUE)

-    contig_tbl = dplyr::mutate(ccdb$contig_tbl, !!sym(cluster_name) := cluster_idx)

-    cluster_tbl = as_tibble(unique(contig_tbl[cluster_name]))

-    replace_cluster_tbl(ccdb, cluster_tbl, contig_tbl, cluster_pk = cluster_name)

+    contig_tbl = ccdb$contig_tbl

+    contig_tbl[[cluster_pk]] = cluster_idx

+    cluster_tbl = as_tibble(unique(contig_tbl[cluster_pk]))

+    replace_cluster_tbl(ccdb, cluster_tbl, contig_tbl, cluster_pk = cluster_pk)

 }

@@ -4,7 +4,7 @@ globalVariables('ngrp')

 #'

 #' @param ccdb [ContigCellDB()]

 #' @param segment_keys fields in `contig_tbl` that identify a cluster

-#' @param cluster_name name of cluster to be added to `cluster_tbl`

+#' @param cluster_pk name of cluster to be added to `cluster_tbl`

 #'

 #' @return [ContigCellDB()]

 #' @export

@@ -13,13 +13,13 @@ globalVariables('ngrp')

 #' data(ccdb_ex)

 #' ccdb_ex = cluster_germline(ccdb_ex)

 #' ccdb_ex$cluster_tbl

-cluster_germline = function(ccdb, segment_keys = c('v_gene', 'j_gene', 'chain'), cluster_name = 'cluster_idx'){

+cluster_germline = function(ccdb, segment_keys = c('v_gene', 'j_gene', 'chain'), cluster_pk = 'cluster_idx'){

     contig_tbl = ccdb$contig_tbl

     seg_types = contig_tbl %>% group_by(!!!syms(segment_keys)) %>% summarize() %>% ungroup()

-    seg_types[[cluster_name]] = seq_len(nrow(seg_types))

+    seg_types[[cluster_pk]] = seq_len(nrow(seg_types))

     cl_con_tbl = left_join_warn(seg_types, contig_tbl, by = segment_keys)

-    cluster_tbl = as_tibble(unique(cl_con_tbl[union(cluster_name, segment_keys)]))

-    replace_cluster_tbl(ccdb, cluster_tbl, cl_con_tbl, cluster_pk = cluster_name)

+    cluster_tbl = as_tibble(unique(cl_con_tbl[union(cluster_pk, segment_keys)]))

+    replace_cluster_tbl(ccdb, cluster_tbl, cl_con_tbl, cluster_pk = cluster_pk)

 }

 globalVariables(c('fc', 'd(medoid)', 'is_medoid', 'n_cluster'))

@@ -8,7 +8,7 @@ cdhit_ccdb(

   ccdb,

   sequence_key,

   type = c("DNA", "AA"),

-  cluster_name = "cluster_idx",

+  cluster_pk = "cluster_idx",

   ...

 )

 }

@@ -19,7 +19,7 @@ cdhit_ccdb(

 \item{type}{one of 'DNA' or 'AA'}

-\item{cluster_name}{\code{character} specifying key, and name for the clustering.}

+\item{cluster_pk}{\code{character} specifying key, and name for the clustering.}

 \item{...}{

   Arguments passed on to \code{\link[=cdhit]{cdhit}}

@@ -7,7 +7,7 @@

 cluster_germline(

   ccdb,

   segment_keys = c("v_gene", "j_gene", "chain"),

-  cluster_name = "cluster_idx"

+  cluster_pk = "cluster_idx"

 )

 }

 \arguments{

@@ -15,7 +15,7 @@ cluster_germline(

 \item{segment_keys}{fields in \code{contig_tbl} that identify a cluster}

-\item{cluster_name}{name of cluster to be added to \code{cluster_tbl}}

+\item{cluster_pk}{name of cluster to be added to \code{cluster_tbl}}

 }

 \value{

 \code{\link[=ContigCellDB]{ContigCellDB()}}

@@ -30,7 +30,7 @@ test_that("Don't need to be sorted", {

 data(ccdb_ex)

 test_that("cdhit_ccdb do update", {

-  res = cdhit_ccdb(ccdb_ex, 'cdr3_nt', type = 'DNA', cluster_name = 'DNA97', identity = .965, min_length = 12, G = 1)

+  res = cdhit_ccdb(ccdb_ex, 'cdr3_nt', type = 'DNA', cluster_pk = 'DNA97', identity = .965, min_length = 12, G = 1)

   expect_is(res, 'ContigCellDB')

   expect_equal(res$cluster_pk, 'DNA97')

   expect_equal(ncol(ccdb_ex$contig_tbl)+1,ncol(res$contig_tbl))

@@ -6,7 +6,7 @@ test_that("Cluster contigs by germline properties",{

   expect_is(cdb,'ContigCellDB')

   expect_gt(ncol(cdb$cluster_tbl),ncol(ccdb_ex$cluster_tbl))

   expect_equal(names(cdb$cluster_tbl),c('cluster_idx','v_gene','j_gene','chain'))

-  cdb1 <- cluster_germline(ccdb_ex,segment_keys = c('v_gene', 'j_gene'),cluster_name = 'clusterID')

+  cdb1 <- cluster_germline(ccdb_ex,segment_keys = c('v_gene', 'j_gene'),cluster_pk = 'clusterID')

   expect_gt(ncol(cdb$cluster_tbl),ncol(cdb1$cluster_tbl))

   expect_equal(names(cdb1$cluster_tbl)[1],'clusterID')

   expect_error(cluster_germline(ccdb_ex,segment_keys = c('v_gene', 'm_gene')))

@@ -15,7 +15,7 @@ test_that("Cluster contigs by germline properties",{

 ccdb_ex_small <- ccdb_ex

 ccdb_ex_small$cell_tbl <- ccdb_ex_small$cell_tbl[1:200,]

 ccdb_ex_small <- cdhit_ccdb(ccdb_ex_small,

-                           sequence_key = 'cdr3_nt', type = 'DNA', cluster_name = 'DNA97',

+                           sequence_key = 'cdr3_nt', type = 'DNA', cluster_pk = 'DNA97',

                            identity = .965, min_length = 12, G = 1)

 ccdb_ex_small_fine <- fine_clustering(ccdb_ex_small, sequence_key = 'cdr3_nt', type = 'DNA')

@@ -23,8 +23,11 @@ library(stringr)

 library(purrr)

 ```

+

 # Load filtered contig files

+We begin with a `data.frame` of concatenated contig files ('all_contig_annotations.csv'), output from the Cellranger VDJ pipeline.

+

 ```{r}

 data(contigs_qc)

 MIN_CDR3_AA = 6

@@ -53,11 +56,9 @@ ggplot(paired_chain, aes(x = interaction(sample, pop), fill = pairing)) + geom_b

 ```{r}

-

-aa80 = cdhit_ccdb(cdb, 'cdr3', type = 'AA', cluster_name = 'aa80', identity = .8)

+aa80 = cdhit_ccdb(cdb, 'cdr3', type = 'AA', cluster_pk = 'aa80', identity = .8)

 aa80 = fine_clustering(aa80, sequence_key = 'cdr3', type = 'AA', keep_clustering_details = TRUE)

-

 # This maybe should be a turned into a function 

 # Other plots should be considered:

 # That show how clusters are split between samples, chains, etc

@@ -71,7 +72,7 @@ We cluster the CDR3 translated amino acid residues with the program [CD-HIT](htt

 # Cluster CDR3 DNA sequences

 ```{r, results = 'hide'}

-cdb = cdhit_ccdb(cdb, 'cdr3_nt', type = 'DNA', cluster_name = 'DNA97', identity = .965, min_length = MIN_CDR3_AA*3-1, G = 1)

+cdb = cdhit_ccdb(cdb, 'cdr3_nt', type = 'DNA', cluster_pk = 'DNA97', identity = .965, min_length = MIN_CDR3_AA*3-1, G = 1)

 cdb = fine_clustering(cdb, sequence_key = 'cdr3_nt', type = 'DNA')

 ggplot(cdb$cluster_tbl %>% filter(n_cluster>1) %>% gather(key, value, -DNA97) , aes(x = value))+ facet_wrap(~key, scales = 'free') + geom_histogram() + scale_y_sqrt()

 ```

@@ -81,7 +82,7 @@ We can also cluster by DNA identity.

 # Cluster by V-J identity

 ```{r}

-germline_cluster = cluster_germline(cdb, segment_keys = c('v_gene', 'j_gene', 'chain'), cluster_name = 'segment_idx')

+germline_cluster = cluster_germline(cdb, segment_keys = c('v_gene', 'j_gene', 'chain'), cluster_pk = 'segment_idx')

 ```

 We can cluster by any other feature of the contigs. Here we cluster each contig based on the chain and V-J genes.  This gives us the set of observed V-J pairings:

@@ -61,7 +61,7 @@ sce = SingleCellExperiment(assay = list(counts = expression), colData = all_bc)

 ```{r}

 ccdb2 = ContigCellDB(ccdb_ex$contig_tbl, contig_pk = ccdb_ex$contig_pk, cell_tbl = colData(sce), cell_pk = ccdb_ex$cell_pk, equalize = FALSE)

-ccdb2 = cdhit_ccdb(ccdb2, 'cdr3', type = 'AA', cluster_name = 'aa80', identity = .8, min_length = 5)

+ccdb2 = cdhit_ccdb(ccdb2, 'cdr3', type = 'AA', cluster_pk = 'aa80', identity = .8, min_length = 5)

 ccdb2 = fine_clustering(ccdb2, sequence_key = 'cdr3', type = 'AA', keep_clustering_details = FALSE)

 ```