V20 ci files

analyses/create-subset-files/01-get_biospecimen_identifiers.R

@@ -57,7 +57,7 @@ get_biospecimen_ids <- function(filename, id_mapping_df) {
     # 'Tumor_Sample_Barcode'
     # if the files have consensus in the name, the first line of the file does
     # not contain MAF version information
-    if (grepl("consensus", filename)) {
+    if (grepl("consensus|hotspots", filename)) {
       snv_file <- data.table::fread(filename, data.table = FALSE)
     } else {
       snv_file <- data.table::fread(filename,

analyses/create-subset-files/02-subset_files.R

@@ -69,7 +69,7 @@ subset_files <- function(filename, biospecimen_ids, output_directory) {
   # filtering strategy depends on the file type, mostly because how the sample
   # IDs change based on the file type -- that's why this logic is required
   if (grepl("pbta-snv", filename)) {
-    if (grepl("consensus-mutation", filename)) {
+    if (grepl("consensus-mutation|hotspots", filename)) {
       snv_file <- data.table::fread(filename, data.table = FALSE)
       snv_file %>%
         dplyr::filter(Tumor_Sample_Barcode %in% biospecimen_ids) %>%

analyses/create-subset-files/create_subset_files.sh

@@ -7,7 +7,7 @@ set -o pipefail
 
 # Set defaults for release and biospecimen file name
 BIOSPECIMEN_FILE=${BIOSPECIMEN_FILE:-biospecimen_ids_for_subset.RDS}
-RELEASE=${RELEASE:-release-v19-20210423}
+RELEASE=${RELEASE:-release-v20-20210726}
 NUM_MATCHED=${NUM_MATCHED:-15}
 
 # This option controls whether or not the two larger MAF files are skipped as
@@ -87,6 +87,9 @@ cp $FULL_DIRECTORY/pbta-mend* $SUBSET_DIRECTORY
 # fusion summary files
 cp $FULL_DIRECTORY/fusion_summary* $SUBSET_DIRECTORY
 
+# MB pathology subtypes
+cp $FULL_DIRECTORY/pbta-mb-pathology-subtypes.tsv $SUBSET_DIRECTORY
+
 # if the md5sum.txt file already exists, get rid of it
 cd $SUBSET_DIRECTORY
 rm -f md5sum.txt