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

 Package: TCGAbiolinks

 Type: Package

 Title: TCGAbiolinks: An R/Bioconductor package for integrative analysis with GDC data

-Version: 2.28.3

+Version: 2.28.4

 Date: 2023-06-06

 Author: Antonio Colaprico,

     Tiago Chedraoui Silva,

@@ -51,20 +51,21 @@ gliomaClassifier <- function(data){

     data(list = models, package = "TCGAbiolinksGUI.data",envir = env)

     for(i in models){

-        model <- get(i,envir = env)

+        message("------------------------------------------------------")

+        message("Model: ",i)

+        model <- get(i, envir = env)

         # If it is a Summarized Experiment object

         # keep only probes used in the model

         aux <- met[,colnames(met) %in% colnames(model$trainingData),drop = FALSE]

-

         # This should not happen!

         if(any(apply(aux,2,function(x) all(is.na(x))))) {

-            print("NA columns")

+            message("o Probes has NA value for all samples. Setting to 0.5 since model does not accept NA")

             aux[,apply(aux,2,function(x) all(is.na(x)))] <- 0.5

         }

         if(any(apply(aux,2,function(x) any(is.na(x))))) {

-            print("NA values")

+            message("o Probes has NA values for some samples. Setting values a the median of the sample since model does not accept NA ")

             colMedians <- colMedians(aux,na.rm = TRUE)

             x <- which(is.na(aux),arr.ind = TRUE)

             for(l in 1:nrow(x)){

@@ -72,6 +73,15 @@ gliomaClassifier <- function(data){

             }

         }

+        # For missing probes add values to 0.5

+        missing_probes <- setdiff(colnames(model$trainingData), colnames(met))

+        if(length(missing_probes) > 0) {

+            message("o Probes are missing. Setting dummy probes to matrix with 0.5 value to all samples.")

+            missing_probes_matrix <- matrix(rep(0.5, nrow(met) * length(missing_probes)),nrow = nrow(met))

+            colnames(missing_probes_matrix) <- missing_probes

+            aux <- bind_cols(aux,missing_probes_matrix)

+        }

+

         pred <- predict(model, aux)

         pred.prob <- predict(model, aux, type = "prob")

         colnames(pred.prob) <- paste0(i,"_", colnames(pred.prob))

@@ -296,7 +296,7 @@ GDCquery_clinic <- function(

             if("treatments" %in% colnames(df)){

                 treatments <- rbindlist(df$treatments,fill = TRUE)

                 df$treatments <- NULL

-                treatments$submitter_id <- gsub("_treatment(_[0-9])?","", treatments$submitter_id)

+                treatments$submitter_id <- gsub("_treatment(_[0-9])?|_treatment([0-9])?","", treatments$submitter_id)

                 treatments <- treatments[,-c("updated_datetime", "state", "created_datetime")]

                 # we have now two types of treatment

@@ -402,7 +402,8 @@ GDCquery_clinic <- function(

             df$project <- project

             df <- df %>% dplyr::relocate(project)

         }

-        if(nrow(results) != nrow(df)){

+

+        if (nrow(results) != nrow(df)) {

             stop("Error: API returned more information")

         }

@@ -139,7 +139,7 @@ GDCprepare <- function(

     }

     cases <- ifelse(

-        grepl("TCGA|TARGET|CGCI-HTMCP-CC",query$results[[1]]$project %>% unlist()),

+        grepl("TCGA|TARGET|CGCI-HTMCP-CC|CPTAC-2",query$results[[1]]$project %>% unlist()),

         query$results[[1]]$cases,

         query$results[[1]]$sample.submitter_id

     )

@@ -478,35 +478,21 @@ readmiRNAIsoformQuantification <- function (files, cases){

     setDF(df)

 }

+

 readSimpleNucleotideVariationMaf <- function(files){

-    ret <- plyr::adply(.data = files,.margins = 1,.fun = function(f){

-        readr::read_tsv(

-            f,

-            comment = "#",

-            col_types = readr::cols(

-                Entrez_Gene_Id = col_integer(),

-                Start_Position = col_integer(),

-                End_Position = col_integer(),

-                t_depth = col_integer(),

-                t_ref_count = col_integer(),

-                t_alt_count = col_integer(),

-                n_depth = col_integer(),

-                TRANSCRIPT_STRAND = col_integer(),

-                PICK = col_integer(),

-                miRNA = col_character(),

-                TSL = col_integer(),

-                HGVS_OFFSET = col_integer()

-            ),

-            progress = TRUE

-        )

-    })

-    if(ncol(ret) == 1) {

-        ret <- plyr::adply(.data = files,.margins = 1,.fun = function(f){

-            read_tsv(

-                f,

+    ret <- files |>

+        purrr::map_dfr(.f = function(x) {

+            tab <- readr::read_tsv(

+                x,

+                show_col_types = FALSE,

                 comment = "#",

-                col_types = cols(

+                col_types = readr::cols(

+                    SOMATIC =  col_character(),

+                    PUBMED = col_character(),

+                    miRNA = col_character(),

+                    HGVS_OFFSET = col_integer(),

+                    PHENO = col_character(),

                     Entrez_Gene_Id = col_integer(),

                     Start_Position = col_integer(),

                     End_Position = col_integer(),

@@ -514,16 +500,18 @@ readSimpleNucleotideVariationMaf <- function(files){

                     t_ref_count = col_integer(),

                     t_alt_count = col_integer(),

                     n_depth = col_integer(),

-                    ALLELE_NUM = col_integer(),

                     TRANSCRIPT_STRAND = col_integer(),

                     PICK = col_integer(),

                     TSL = col_integer(),

-                    HGVS_OFFSET = col_integer(),

-                    MINIMISED = col_integer()),

-                progress = TRUE

-            )

+                    DISTANCE = col_integer()

+                ))

+

+            # empty MAF file

+            # https://portal.gdc.cancer.gov/files/7917fcbe-cb66-447d-8ea8-3a324feee3fa

+            if(nrow(tab) == 0) {return (NULL)}

+            tab

         })

-    }

+

     return(ret)

 }

@@ -55,25 +55,25 @@ the same result as the paper.

 ```{r, eval = FALSE, message = FALSE, results = "hide"}

 query <- GDCquery(

-  project = "TCGA-GBM",

-  data.category = "DNA methylation",

-  barcode = c("TCGA-06-0122","TCGA-14-1456"),

-  platform = "Illumina Human Methylation 27",

-  legacy = TRUE

+    project = "TCGA-GBM",

+    data.category = "DNA Methylation",

+    barcode = c("TCGA-06-0122","TCGA-14-1456"),

+    platform = "Illumina Human Methylation 27",

+    data.type = "Methylation Beta Value"

 )

 GDCdownload(query)

-data.hg19 <- GDCprepare(query)

+dnam <- GDCprepare(query)

 ```

 ```{r, eval = FALSE}

-assay(data.hg19)[1:5,1:2]

+assay(dnam)[1:5,1:2]

 ```

 ## Function

 <hr>

 ```{r, eval = FALSE}

-classification <- gliomaClassifier(data.hg19)

+classification <- gliomaClassifier(dnam)

 ```

 ## Results