@@ -1,8 +1,8 @@

 Package: OncoSimulR

 Type: Package

 Title: Forward Genetic Simulation of Cancer Progression with Epistasis 

-Version: 2.9.3

-Date: 2017-11-27

+Version: 2.9.6

+Date: 2017-12-27

 Authors@R: c(person("Ramon", "Diaz-Uriarte", role = c("aut", "cre"),

 		     email = "rdiaz02@gmail.com"),

 	      person("Mark", "Taylor", role = "ctb", email = "ningkiling@gmail.com"))

@@ -327,6 +327,28 @@ samplePop <- function(x, timeSample = "last",

 }

+

+## single_largest_last_pop <- function(x) {

+##     last <- nrow(x$pops.by.time)

+##     largest <- which.max(x$pops.by.time[last, , drop = FALSE]) - 1

+##     genot <- x[["GenotypesLabels"]][largest]

+##     strsplit(genot, split = ", ")[[1]]

+## }

+

+

+## ## like samplePop(x, timeSample = "last") but return the single most abundant genotype

+## Just a prototype. Not well tested.

+## largest_last_pop <- function(x) {

+##     y <- lapply(x, single_largest_last_pop)

+##     allg <- sort(unique(unlist(y)))

+##     m <- t(vapply(y, function(z) {as.integer(allg %in%  z) },

+##                   integer(length(allg)) ))

+##     colnames(m) <- allg

+##     return(m)

+## }

+

+

+

 oncoSimulPop <- function(Nindiv,

                          fp,

                          model = "Exp",

@@ -767,8 +789,39 @@ print.oncosimul <- function(x, ...) {

 ## }

 summary.oncosimulpop <- function(object, ...) {

+    ## some simulations could have failed seriously returning a NULL

+    ## FIXME: how, why? Out of memory?

+

+    rm1 <- which(unlist(lapply(object, is.null)))

+    if(length(rm1) > 0) {

+        if(length(rm1) < length(object)) {

+            warning("Some simulations returned NULL. They will be removed",

+                    " from the summary. The NULL runs are ",

+                    paste(rm1, collapse = ", "),

+                    ".")

+        } else {

+            warning("All simulations returned NULL.")

+            return(NA)

+        }

+    }

+

+    ## I do not want to rm above for two reasons:

+    ##   - avoid re-asigning to the object

+    ##   - changing the numbering of the indices below if NULLs

+    ## So I need something more involved

+    ## Figure out exactly what the summary of a NULL is

+    sumnull <- summary(NULL)

+    

     tmp <- lapply(object, summary)

-    rm <- which(unlist(lapply(tmp, function(x) (length(x) == 1) && (is.na(x)))))

+

+    ## rm <- which(unlist(lapply(tmp,

+    ##                           function(x) (length(x) == 1) &&

+    ##                                       (is.na(x) || is.null(x)))))

+

+    rm <- which(unlist(lapply(tmp,

+                              function(x) ((length(x) == 1) &&

+                                           (is.na(x) || is.null(x))) ||

+                                          (identical(x, sumnull)))))

     if(length(rm) > 0)

         if(length(rm) < length(object)) {

         warning("Some simulations seem to have failed and will be removed",

@@ -1631,14 +1684,58 @@ get.mut.vector <- function(x, timeSample, typeSample,

         return( as.numeric((tcrossprod(pop,

                                        x$Genotypes)/popSize) >= thresholdWhole) )

     } else if (typeSample %in%  c("singleCell", "single")) {

-

         return(x$Genotypes[, sample(seq_along(pop), 1, prob = pop)])

+    } else if (typeSample %in% c("singleCell-noWT", "single-noWT",

+                                 "singleCell-nowt", "single-nowt")) {

+        genots <- x$Genotypes

+        whichwt <- which(x$GenotypesLabels == "")

+        if(length(whichwt)) {

+            genots <- genots[, -whichwt, drop = FALSE]

+            pop <- pop[-whichwt]

+        }

+        if(all(pop == 0)) {

+            warning("No non-WT clone with required popSize or at required time")

+            return(rep(NA, nrow(x$Genotypes)))

+        } else {

+            return(genots[, sample(seq_along(pop), 1, prob = pop)])

+        }

     } else {

         stop("Unknown typeSample option")

     }

 }

+## get.mut.vector <- function(x, timeSample, typeSample,

+##                            thresholdWhole, popSizeSample) {

+##     if(is.null(x$TotalPopSize)) {

+##         warning(paste("It looks like this simulation never completed.",

+##                       " Maybe it reached maximum allowed limits.",

+##                       " You will get NAs"))

+##         return(rep(NA, length(x$geneNames)))

+##     }

+##     the.time <- get.the.time.for.sample(x, timeSample, popSizeSample)

+##     if(the.time < 0) { 

+##         return(rep(NA, nrow(x$Genotypes)))

+##     } 

+##     pop <- x$pops.by.time[the.time, -1]

+    

+##     if(all(pop == 0)) {

+##         stop("You found a bug: this should never happen")

+##     }

+    

+##     if(typeSample %in% c("wholeTumor", "whole")) {

+##         popSize <- x$PerSampleStats[the.time, 1]

+##         return( as.numeric((tcrossprod(pop,

+##                                        x$Genotypes)/popSize) >= thresholdWhole) )

+##     } else if (typeSample %in%  c("singleCell", "single")) {

+

+##         return(x$Genotypes[, sample(seq_along(pop), 1, prob = pop)])

+##     } else {

+##         stop("Unknown typeSample option")

+##     }

+## }

+

+

@@ -1,7 +1,8 @@

 citHeader("If you use OncoSimulR, please cite the OncoSimulR Bioinformatics paper.",

-          " A former version of OncoSimulR has been used in a large",

-          " comparative study of methods to infer restrictions,",

-          " published in BMC Bioinformatics; you might want to cite that too,",

+          " OncoSimulR has been used in two large",

+          " comparative studies of methods to infer restrictions in",

+          " the order of accumulation of mutations (cancer progression models)",

+          " published in Bioinformatics and BMC Bioinformatics; you might want to cite those too,",

           " if appropriate, such as when referring to using evolutionary simulations",

           " to assess oncogenetic tree/cancer progression methods performance.")

@@ -11,26 +12,38 @@ citEntry(entry="Article",

          title = "OncoSimulR: genetic simulation with arbitrary epistasis and mutator genes in asexual populations.",

          journal = "Bioinformatics",

          year = "2017",

+         volume = "33",

+         number = "12",

+         pages = "1898--1899",

          doi = "10.1093/bioinformatics/btx077",

          publisher = "Oxford University Press",

          url = " https://doi.org/10.1093/bioinformatics/btx077",

          textVersion = paste("R Diaz-Uriarte.",

                              "OncoSimulR: genetic simulation with arbitrary epistasis and mutator genes in asexual populations.",

-                             " 2017. Bioinformatics, https://doi.org/10.1093/bioinformatics/btx077.")

+                             " 2017. Bioinformatics, 33, 1898--1899. https://doi.org/10.1093/bioinformatics/btx077.")

          )

-## citEntry(entry="Manual",

-##          author = "R Diaz-Uriarte",

-##          title = "OncoSimulR: Forward Genetic Simulation of Cancer Progresion with Epistasis.",

-##          year = "2015",

-##          note = "R package version 2.0.0",

-##          textVersion = paste("R Diaz-Uriarte.",

-##              "OncoSimulR: Forward Genetic Simulation of Cancer Progresion with Epistasis. 2015. BioConductor package version 2.0.0 ")

-## )

 ## citHeader("A former version of OncoSimulR has been used in this paper:")

+

+citEntry(entry="Article",

+         author = "R Diaz-Uriarte",

+         title = "Cancer progression models and fitness landscapes: a many-to-many relationship",

+         journal = "Bioinformatics",

+         year = "2017",

+         doi = "10.1093/bioinformatics/btx663",

+         url = "https://academic.oup.com/bioinformatics/advance-article/doi/10.1093/bioinformatics/btx663/",

+##         volume = "zz",

+##         number = "zz",

+         textVersion = paste("R Diaz-Uriarte.",

+                             "Cancer progression models and fitness landscapes: a many-to-many relationship",

+                             "2017", "Bioinformatics.",

+                             "https://academic.oup.com/bioinformatics/advance-article/doi/10.1093/bioinformatics/btx663/")

+)

+

+

 citEntry(entry="Article",

          author = "R Diaz-Uriarte",

          title = "Identifying restrictions in the order of accumulation of mutations during tumor progression: effects of passengers, evolutionary models, and sampling",

@@ -1,3 +1,16 @@

+Changes in version 2.9.6 (2017-12-27):

+	- Updated citation.

+	- An example (in miscell-files) about using and stopping with

+	modules.

+	- Prototype for sampling the single larges pop at last period

+	(function largest_last_pop, commented out for now).

+

+Changes in version 2.9.5 (2017-12-7):

+	- samplePop: new option "single-nowt"

+

+Changes in version 2.9.4 (2017-11-30):

+	- Deal with the very rare NULL simulations in summary.

+

 Changes in version 2.9.3 (2017-11-27):

 	- Make clang happy (do not use flandscape as DataFrame)

@@ -57,6 +57,8 @@ sampledGenotypes(y, genes = NULL)

     probability of sampling a cell (a clone) is directly proportional to

     its population size.  "wholeTumor" (or "whole") for whole tumor

     sampling (i.e., this is similar to a biopsy being the entire tumor).

+    "singleCell-noWT" or "single-nowt" is single cell sampling, but

+    excluding the wild type.

   }

   \item{thresholdWhole}{

@@ -252,7 +252,13 @@ test_that("exercising the sampling code, v2 objects", {

               expect_message(samplePop(o4, timeSample = "last",

                                        typeSample = "whole"),

                              "Subjects by Genes matrix of 2 subjects and 10 genes")

-          })

+              expect_message(samplePop(o4, typeSample = "single-nowt",

+                                       timeSample = "last"),

+                            "Subjects by Genes matrix of 2 subjects and 10 genes")

+              expect_message(samplePop(o4, typeSample = "single-nowt",

+                                       timeSample = "uniform"),

+                             "Subjects by Genes matrix of 2 subjects and 10 genes")

+})

 test_that("exercising the sampling code, v2 objects, more", {

               cs <-  data.frame(parent = c(rep("Root", 4), "a", "b", "d", "e", "c"),

@@ -300,7 +306,13 @@ test_that("exercising the sampling code, v2 objects, more", {

               expect_message(samplePop(o4, timeSample = "last",

                                        typeSample = "whole"),

                              "Subjects by Genes matrix of 4 subjects and 6 genes")

-          })

+              expect_message(samplePop(o4, typeSample = "single-nowt",

+                                       timeSample = "last"),

+                            "Subjects by Genes matrix of 4 subjects and 6 genes")

+              expect_message(samplePop(o4, typeSample = "single-nowt",

+                                       timeSample = "uniform"),

+                             "Subjects by Genes matrix of 4 subjects and 6 genes")

+})

@@ -382,7 +394,37 @@ test_that("exercising sampling code, customSize", {

     expect_message(samplePop(o4, typeSample = "single",

                              popSizeSample = c(6100, 0, 5000, 9000)),

                    "Subjects by Genes matrix of 4 subjects and 6 genes")

-   

+

+    expect_message(samplePop(o4, typeSample = "single-nowt",

+                             popSizeSample = c(9000, 9000, 8500, 9000)),

+                   "Subjects by Genes matrix of 4 subjects and 6 genes")

+

+    ## these are specific for the nowt code to exercise the dealing with

+    ## borderline cases and the second to deal with a clearly non-borderline

+    o41 <- oncoSimulPop(4,

+                        cbn1,

+                        initSize = 2e3,

+                       detectionSize = 1e3,

+                       onlyCancer = TRUE,

+                       mc.cores = 2,

+                       max.num.tries = 5000,

+                       sampleEvery = 0.03, keepEvery = 1)

+    expect_message(samplePop(o41, typeSample = "single-nowt",

+                             popSizeSample = c(900, 800, 850, 900)),

+                   "Subjects by Genes matrix of 4 subjects and 6 genes")

+    expect_warning(samplePop(o41, typeSample = "single-nowt",

+                             popSizeSample = c(900, 800, 850, 900)),

+                   "No non-WT clone with required popSize or at required time")

+    o91 <- oncoSimulPop(4,

+                       cbn1, 

+                       detectionSize = 1e5,

+                       onlyCancer = TRUE,

+                       mc.cores = 2,

+                       max.num.tries = 5000,

+                       sampleEvery = 0.03, keepEvery = 1)

+     expect_message(samplePop(o91, typeSample = "single-nowt",

+                             popSizeSample = c(9000, 9000, 8500, 9000)),

+                   "Subjects by Genes matrix of 4 subjects and 6 genes")

 })

@@ -1,15 +1,15 @@

 \usepackage[%

-		shash={8f168d4},

-		lhash={8f168d4bffbf847d6faff5e0485705fe32878aa0},

-		authname={Ramon Diaz-Uriarte (at Coleonyx)},

+		shash={7c279ec},

+		lhash={7c279ec4db10924e57799b5d15e9d89217b7c785},

+		authname={Ramon Diaz-Uriarte (at Draco)},

 		authemail={rdiaz02@gmail.com},

-		authsdate={2017-11-14},

-		authidate={2017-11-14 16:12:04 +0100},

-		authudate={1510672324},

-		commname={Ramon Diaz-Uriarte (at Coleonyx)},

+		authsdate={2017-12-01},

+		authidate={2017-12-01 14:30:15 +0100},

+		authudate={1512135015},

+		commname={Ramon Diaz-Uriarte (at Draco)},

 		commemail={rdiaz02@gmail.com},

-		commsdate={2017-11-14},

-		commidate={2017-11-14 16:12:04 +0100},

-		commudate={1510672324},

+		commsdate={2017-12-01},

+		commidate={2017-12-01 14:30:15 +0100},

+		commudate={1512135015},

 		refnames={ (HEAD -> master, origin/master, origin/HEAD)}

 	]{gitsetinfo}

\ No newline at end of file