v12 hgg subtyping (10/N)

.github/workflows/continuous_integration.yml

@@ -66,12 +66,10 @@ jobs:
 
           - name: Molecular Subtyping - EPN
             entrypoint: molecular-subtyping-EPN/run-molecular-subtyping-EPN.sh
-            openpbta_subset: 0
 
           - name: Molecular Subtyping - EMBRYONAL
             entrypoint: molecular-subtyping-embryonal/run-embryonal-subtyping.sh
             openpbta_testing: 1
-            openpbta_subset: 0
 
           - name: Molecular Subtyping - CHORDOMA
             entrypoint: molecular-subtyping-chordoma/run-molecular-subtyping-chordoma.sh
@@ -85,7 +83,6 @@ jobs:
 
           - name: Molecular Subtyping - HGG
             entrypoint: molecular-subtyping-HGG/run-molecular-subtyping-HGG.sh
-            openpbta_subset: 0
 
           - name: Molecular Subtyping - LGG
             entrypoint: molecular-subtyping-LGAT/run_subtyping.sh
@@ -134,8 +131,8 @@ jobs:
           - name: TMB calculation
             entrypoint: tmb-calculation/run_tmb_calculation.sh
 
-          - name: Immune Deconvolution
-            entrypoint: immune-deconv/run-immune-deconv.sh
+          #- name: Immune Deconvolution
+          #  entrypoint: immune-deconv/run-immune-deconv.sh
 
           #- name: EFO/MONDO annotation
           #  entrypoint: efo-mondo-mapping/run_search_and_qc.sh

Dockerfile

@@ -13,6 +13,9 @@ COPY scripts/install_bioc.r .
 
 ### Install apt-getable packages to start
 #########################################
+
+# stretch is EOL, so we need to use the archive
+RUN echo "deb http://archive.debian.org/debian stretch main" > /etc/apt/sources.list
 RUN apt-get update && apt-get install -y --no-install-recommends apt-utils dialog
 
 # Add curl, bzip2 and some dev libs

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

@@ -79,7 +79,9 @@ get_biospecimen_ids <- function(filename, id_mapping_df) {
     biospecimen_ids <- unique(bed_file$Kids_First_Biospecimen_ID)
   } else if (grepl("cnv", filename)) {
     # the two CNV files now have different structures
-    cnv_file <- readr::read_tsv(filename)
+    if (stringr::str_detect(filename, "gistic", negate = TRUE)) {
+      cnv_file <- readr::read_tsv(filename)
+    }
     if (grepl("controlfreec|cnvkit_with_status", filename)) {
       biospecimen_ids <- unique(cnv_file$Kids_First_Biospecimen_ID)
     } else if (grepl("consensus_wgs_plus_cnvkit_wxs", filename)) {
@@ -96,10 +98,8 @@ get_biospecimen_ids <- function(filename, id_mapping_df) {
     fusion_file <- readr::read_tsv(filename)
     # the biospecimen IDs in the filtered/prioritize fusion list included with
     # the download are in a column called 'Sample'
-    if (grepl("putative-oncogenic", filename)) {
+    if (grepl("putative-oncogenic|dgd|annoFuse", filename)) {
       biospecimen_ids <- unique(fusion_file$Sample)
-    } else if(grepl("dgd", filename)) {
-      biospecimen_ids <- unique(fusion_file$Tumor_Sample_Barcode)
     } else if (grepl("fusion_summary", filename)) {
       biospecimen_ids <- unique(fusion_file$Kids_First_Biospecimen_ID)
     } else {
@@ -502,7 +502,7 @@ matched_participant_id_list <- purrr::map(
 nonmatched_participant_id_list <-
   purrr::map(participant_id_list,
              ~ setdiff(.x, matched_participant_id_list)) %>%
-  purrr::map(~ sample(.x, num_nonmatched_participants))
+  purrr::map(~ sample(.x, min(length(.x),num_nonmatched_participants)))
 
 # combine matched and nonmatched lists of ids for subsetting
 participant_ids_for_subset <- 

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

@@ -74,29 +74,35 @@ 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("snv", filename)) {
-    if (grepl("hotspots", filename)) {
-      snv_file <- data.table::fread(filename,
-                                    skip = 1,  # skip version string
-                                    data.table = FALSE,
-                                    showProgress = FALSE)
-      # we need to obtain the version string from the first line of the MAF file
-      version_string <- readLines(filename, n = 1)
-      # filter + write to file with custom function
-      snv_file %>%
-        dplyr::filter(Tumor_Sample_Barcode %in% biospecimen_ids) %>%
-        write_maf_file(file_name = output_file,
-                       version_string = version_string)
-      snv_file %>%
-        dplyr::filter(Tumor_Sample_Barcode %in% biospecimen_ids) %>%
-        readr::write_tsv(output_file)
-    } else {
-      # in a column 'Tumor_Sample_Barcode'
-      snv_file <- data.table::fread(filename, data.table = FALSE, 
-                                    showProgress = FALSE)
-      snv_file %>%
-        dplyr::filter(Tumor_Sample_Barcode %in% biospecimen_ids) %>%
-        readr::write_tsv(output_file)
-    }
+    # if (grepl("hotspots", filename)) {
+    #   snv_file <- data.table::fread(filename,
+    #                                 skip = 1,  # skip version string
+    #                                 data.table = FALSE,
+    #                                 showProgress = FALSE)
+    #   # we need to obtain the version string from the first line of the MAF file
+    #   version_string <- readLines(filename, n = 1)
+    #   # filter + write to file with custom function
+    #   snv_file %>%
+    #     dplyr::filter(Tumor_Sample_Barcode %in% biospecimen_ids) %>%
+    #     write_maf_file(file_name = output_file,
+    #                    version_string = version_string)
+    #   snv_file %>%
+    #     dplyr::filter(Tumor_Sample_Barcode %in% biospecimen_ids) %>%
+    #     readr::write_tsv(output_file)
+    # } else {
+    #   # in a column 'Tumor_Sample_Barcode'
+    #   snv_file <- data.table::fread(filename, data.table = FALSE, 
+    #                                 showProgress = FALSE)
+    #   snv_file %>%
+    #     dplyr::filter(Tumor_Sample_Barcode %in% biospecimen_ids) %>%
+    #     readr::write_tsv(output_file)
+    # }
+    # in a column 'Tumor_Sample_Barcode'
+    snv_file <- data.table::fread(filename, data.table = FALSE, 
+                                  showProgress = FALSE)
+    snv_file %>% 
+      dplyr::filter(Tumor_Sample_Barcode %in% biospecimen_ids) %>%
+      readr::write_tsv(output_file)
   } else if (grepl("biospecimen", filename)) {
     # in a column 'Kids_First_Biospecimen_ID'
     bed_file <- readr::read_tsv(filename)
@@ -130,8 +136,12 @@ subset_files <- function(filename, biospecimen_ids, output_directory) {
         readr::write_tsv(output_file)
     } else if (grepl("dgd", filename)) {
       fusion_file %>%
-        dplyr::filter(Tumor_Sample_Barcode %in% biospecimen_ids) %>%
+        dplyr::filter(Sample %in% biospecimen_ids) %>%
         readr::write_tsv(output_file)
+    } else if (grepl("annoFuse", filename)) {
+        fusion_file %>%
+	dplyr::filter(Sample %in% biospecimen_ids) %>%
+	readr::write_tsv(output_file)
     } else if (grepl("fusion_summary", filename)) {
       fusion_file %>%
         dplyr::filter(Kids_First_Biospecimen_ID %in% biospecimen_ids) %>%

analyses/create-subset-files/create_subset_files.sh

@@ -8,7 +8,7 @@ set -o pipefail
 
 # Set defaults for release and biospecimen file name
 BIOSPECIMEN_FILE=${BIOSPECIMEN_FILE:-biospecimen_ids_for_subset.RDS}
-RELEASE=${RELEASE:-v11}
+RELEASE=${RELEASE:-v12}
 NUM_MATCHED=${NUM_MATCHED:-15}
 
 # This option controls whether or not the two larger MAF files are skipped as
@@ -76,9 +76,7 @@ cp $FULL_DIRECTORY/histologies-base.tsv $SUBSET_DIRECTORY
 cp $FULL_DIRECTORY/uberon-map-gtex-*.tsv $SUBSET_DIRECTORY
 cp $FULL_DIRECTORY/efo-mondo-map.tsv $SUBSET_DIRECTORY
 cp $FULL_DIRECTORY/ensg-hugo-pmtl-mapping.tsv $SUBSET_DIRECTORY
-cp $FULL_DIRECTORY/infinium-annotation-mapping.tsv $SUBSET_DIRECTORY
-cp $FULL_DIRECTORY/infinium-methylationepic-v-1-0-b5-manifest-file-csv.zip $SUBSET_DIRECTORY
-cp $FULL_DIRECTORY/UCSC_hg19-GRCh37_Ensembl2RefSeq.tsv $SUBSET_DIRECTORY
+cp $FULL_DIRECTORY/infinium.gencode.v39.probe.annotations.tsv.gz $SUBSET_DIRECTORY
 
 # GISTIC output
 cp $FULL_DIRECTORY/cnv-consensus-gistic.zip $SUBSET_DIRECTORY

analyses/efo-mondo-mapping/results/efo-mondo-map-prefill.tsv

@@ -6,6 +6,9 @@ Adenocarcinoma	EFO_0000228	MONDO_0004970	NCIT_C2852
 Adrenocortical Carcinoma	EFO_1000796	MONDO_0006639	NCIT_C9325
 Anaplastic Large Cell Lymphoma	EFO_0003032	MONDO_0020325	NCIT_C3720
 Angiosarcoma	EFO_0003968	MONDO_0016982	NCIT_C3088
+Astroblastoma	NA	NA	NA
+Astrocytoma	NA	NA	NA
+Atypical choroid plexus papilloma	NA	NA	NA
 Atypical Teratoid Rhabdoid Tumor	EFO_1002008	MONDO_0020560	NCIT_C6906
 B Acute Lymphoblastic Leukemia/Lymphoma	EFO_0000094	MONDO_0004967	NCIT_C8644
 Bladder Urothelial Carcinoma	EFO_0006544	MONDO_0005611	NCIT_C39851
@@ -15,58 +18,60 @@ Cavernoma	EFO_1000151	MONDO_0003155	NCIT_C3086
 Central neurocytoma	EFO_1000856	MONDO_0019134	NCIT_C3791
 Cervical Squamous Cell Carcinoma and Endocervical Adenocarcinoma	EFO_1000162	MONDO_0006143	NCIT_C157526
 Cholangiocarcinoma	EFO_0005221	MONDO_0019087	NCIT_C4436
+Chondromyxoid fibroma	EFO_0000332	MONDO_0018447	NCIT_C3830
 Chordoma	Orphanet_178	MONDO_0008978	NCIT_C2947
 Choroid plexus carcinoma	MONDO_0016718	MONDO_0016718	NCIT_C4715
+Choroid plexus papilloma	EFO_1000177	MONDO_0009837	NCIT_C3698
 Choroid plexus tumor	EFO_0007206	MONDO_0016717	NCIT_C4533
 Chromophobe renal cell carcinoma	EFO_0000335	MONDO_0017885	NCIT_C4146
 Chronic Myelogenous Leukemia	EFO_0000339	MONDO_0011996	NCIT_C3174
 CIC-DUX4 Sarcoma	EFO_0000691	MONDO_0005089	NCIT_C165663
 Clear cell sarcoma of the kidney	EFO_0000350	MONDO_0005006	NCIT_C4264
-CNS Burkitt's lymphoma	EFO_1000157	Orphanet_46135	NCIT_C9189
+CNS Burkitt's lymphoma	NA	NA	NA
 CNS Embryonal tumor	EFO_0005784	MONDO_0018843	NCIT_C5398
 CNS Melanoma	EFO_0002617	MONDO_0005191	NCIT_C133504
 CNS neuroblastoma	EFO_0000621	MONDO_0006130	NCIT_C4826
 Colon Adenocarcinoma	EFO_1001949	MONDO_0002271	NCIT_C4349
 Craniopharyngioma	EFO_1000209	MONDO_0002787	NCIT_C2964
 Cutaneous Melanoma	EFO_0000389	MONDO_0005012	NCIT_C3510
-Desmoid-type fibromatosis	EFO_0009907	Orphanet_873	NCIT_C9182
 Desmoplastic infantile astrocytoma and ganglioglioma	MONDO_0016731	MONDO_0016731	NCIT_C4747
-Diffuse fibrillary astrocytoma	Orphanet_251601	MONDO_0016688	NCIT_C4322
+Diffuse fibrillary astrocytoma	NA	NA	NA
+Diffuse hemispheric glioma	MONDO_0016680	MONDO_0016680	NA
 Diffuse intrinsic pontine glioma	EFO_1000026	MONDO_0006033	NCIT_C94764
 Diffuse leptomeningeal glioneuronal tumor	MONDO_0016745	MONDO_0016745	NCIT_C129424
 Diffuse midline glioma	EFO_1000026	MONDO_0006033	NCIT_C129309
 Dysembryoplastic neuroepithelial tumor	EFO_0005551	MONDO_0005505	NCIT_C9505
 Dysgerminoma	MONDO_0003002	MONDO_0003002	NCIT_C2996
-Embryonal tumor	EFO_0005784	MONDO_0005564	NCIT_C3264
-Embryonal Tumor NOS	EFO_0005784	MONDO_0005564	NCIT_C5398
 Embryonal tumor with multilayer rosettes	MONDO_0016715	MONDO_0016715	NCIT_C129499
 Ependymoma	EFO_1000028	MONDO_0016698	NCIT_C3017
+Epstein-Barr virus-related tumor	MONDO_0017342	MONDO_0017342	NA
 Esophageal Carcinoma	EFO_0002916	MONDO_0019086	NCIT_C3513
 Ewing sarcoma	EFO_0000174	MONDO_0012817	NCIT_C4817
-Extraventricular neurocytoma	EFO_1000856	MONDO_0016727	NCIT_C92555
+Extraventricular neurocytoma	MONDO_0016727	MONDO_0016727	NCIT_C92555
 Fibromyxoid lesion	MONDO_0037745	MONDO_0037745	NCIT_C66760
 Ganglioglioma	EFO_0003094	MONDO_0016733	NCIT_C3788
 Ganglioneuroblastoma	EFO_0000502	MONDO_0005035	NCIT_C3790
 Ganglioneuroma	EFO_0000500	MONDO_0005033	NCIT_C3049
 Germ Cell Tumor	EFO_0000514	MONDO_0005040	NCIT_C3708
 Germinoma	MONDO_0020580	MONDO_0020580	NCIT_C121618
-Glial-neuronal tumor NOS	MONDO_0016729	MONDO_0016729	NCIT_C4747
+Glial-neuronal tumor	MONDO_0016729	MONDO_0016729	NCIT_C4747
+Glioblastoma	NA	NA	NA
 Glioblastoma Multiforme	EFO_0000519	MONDO_0018177	NCIT_C3058
-Gliomatosis Cerebri	MONDO_0016683	MONDO_0016683	NCIT_C4318
-Gliosarcoma	EFO_1001465	MONDO_0016681	NCIT_C3796
 Head and Neck Squamous Cell Carcinoma	EFO_0000181	MONDO_0010150	NCIT_C34447
 Hemangioblastoma	MONDO_0016748	MONDO_0016748	NCIT_C3801
 Hepatoblastoma	EFO_1000292	MONDO_0018666	NCIT_C3728
 Hepatocellular Carcinoma	EFO_0000182	MONDO_0007256	NCIT_C3099
-High-grade glioma/astrocytoma	MONDO_0016680	MONDO_0016680	NCIT_C102897
+High-grade glioma	MONDO_0100342	MONDO_0100342	NCIT_C4822
 Histiocytic tumor	MONDO_0020081	MONDO_0020081	NCIT_C9294
 Hodgkin's lymphoma	EFO_0000183	MONDO_0004952	NCIT_C9357
+Infant-type hemispheric glioma	EFO_0005543	MONDO_0014695	NCIT_C185471
 Infantile Fibrosarcoma	MONDO_0002678	MONDO_0002678	NCIT_C4244
 Inflammatory Myofibroblastic Tumor	MONDO_0015798	MONDO_0015798	NCIT_C6481
 Intrahepatic Cholangiocarcinoma	EFO_1001961	MONDO_0003210	NCIT_C35417
+Intraneural perineuroma	MONDO_0015032	MONDO_0015032	NCIT_C6911
 Juvenile xanthogranuloma	EFO_1000311	MONDO_0015534	NCIT_C3451
 Langerhans Cell histiocytosis	EFO_1000318	MONDO_0018310	NCIT_C3107
-Low-grade glioma/astrocytoma	MONDO_0016685	MONDO_0016685	NCIT_C116342
+Low-grade glioma	MONDO_0021637	MONDO_0021637	NCIT_C132067
 Lung Adenocarcinoma	EFO_0000571	MONDO_0005061	NCIT_C3512
 Lung Squamous Cell Carcinoma	EFO_0000708	MONDO_0005097	NCIT_C3493
 Lymphoid Neoplasm Diffuse Large B-cell Lymphoma	EFO_0000403	MONDO_0018905	NCIT_C8851
@@ -75,24 +80,25 @@ Medulloblastoma	EFO_0002939	MONDO_0007959	NCIT_C3222
 Melanocytic tumor	MONDO_0003222	MONDO_0003222	NCIT_C5504
 Melanoma	EFO_0000756	MONDO_0005105	NCIT_C3224
 Meningioma	Orphanet_2495	MONDO_0016642	NCIT_C3230
+Mesenchymal tumor	EFO_1000473	MONDO_0003512	NCIT_C7059
 Mesothelioma	EFO_0000588	MONDO_0005065	NCIT_C3234
 Metastatic secondary tumors	EFO_0009812	MONDO_0024883	NCIT_C4968
 Mixed germ cell tumor	MONDO_0015864	MONDO_0015864	NCIT_C4290
-Myeloid Neoplasm	EFO_0002427	MONDO_0005170	NCIT_C9290
 Neuroblastoma	EFO_0000621	MONDO_0005072	NCIT_C3270
-Neuroepithelial neoplasm	MONDO_0021193	MONDO_0021193	NCIT_C3787
 Neurofibroma/Plexiform	EFO_0000658	MONDO_0003304	NCIT_C3797
+Non-germinomatous germ cell tumor	MONDO_0020580	MONDO_0020580	NCIT_C121619
 Non-Hodgkin Lymphoma	EFO_0005952	MONDO_0018908	NCIT_C3211
 Oligodendroglioma	EFO_0000632	MONDO_0016695	NCIT_C3288
 Osteosarcoma	EFO_0000637	MONDO_0009807	NCIT_C9145
 Ovarian Serous Cystadenocarcinoma	EFO_1000043	MONDO_0006046	NCIT_C7978
 Pancreatic Adenocarcinoma	EFO_1000044	MONDO_0006047	NCIT_C8294
+Perineuroma	MONDO_0019404	MONDO_0019404	NCIT_C4973
 Pheochromocytoma and Paraganglioma	EFO_0020005	MONDO_0035540	NA
-Pilocytic astrocytoma	MONDO_0016691	MONDO_0016691	NCIT_C4047
-Pineal tumor	MONDO_0021232	MONDO_0021232	NCIT_C3328
+Pilocytic astrocytoma	NA	NA	NA
 Pineoblastoma	EFO_1000475	MONDO_0016722	NCIT_C9344
 Pineocytoma	EFO_1000476	MONDO_0016723	NCIT_C6966
 Pleomorphic xanthoastrocytoma	MONDO_0016690	MONDO_0016690	NCIT_C4323
+Primary mediastinal large B cell lymphoma	MONDO_0004021	MONDO_0020323	NCIT_C9280
 Prostate Adenocarcinoma	EFO_0000673	MONDO_0005082	NCIT_C2919
 Rectum Adenocarcinoma	EFO_0005631	MONDO_0002169	NCIT_C9383
 Renal Clear Cell Carcinoma	EFO_0000349	MONDO_0005005	NCIT_C4033
@@ -116,4 +122,4 @@ Thyroid Gland Papillary Carcinoma	EFO_0000641	MONDO_0005075	NCIT_C4035
 Uterine Carcinosarcoma	EFO_1000613	MONDO_0006485	NCIT_C42700
 Uterine Corpus Endometrial Carcinoma	EFO_0007532	MONDO_0000553	NCIT_C159413
 Uveal Melanoma	EFO_1000616	MONDO_0006486	NCIT_C7712
-Wilms tumor	MONDO_0006058	MONDO_0006058	NCIT_C3267
+Wilms tumor	EFO_1000056	MONDO_0006058	NCIT_C3267