CIVIC, NCI Thesaurus, MitelmanDB and Depmap update

sigven · sigven · commit b94f2e79483d · 2024-08-07T12:42:54.000+02:00
diff --git a/DESCRIPTION b/DESCRIPTION
@@ -2,6 +2,7 @@ Package: pharmOncoX
 Type: Package
 Title: Molecularly targeted cancer drugs and biomarkers
 Version: 1.6.10
+URL: https://sigven.github.io/pharmOncoX
 Authors@R: 
     c(person(given = "Sigve",
              family = "Nakken",
@@ -19,7 +20,6 @@ Description: This data package collects anticancer drug information from
     also allows for the retrieval of curated biomarkers from multiple 
     freely available resources (CIViC, CGI, Mitelman database).
 License: MIT + file LICENSE
-URL: https://github.com/sigven/pharmOncoX
 BugReports: https://github.com/sigven/pharmOncoX/issues
 Depends: 
     R (>= 4.1.0)
diff --git a/NEWS.md b/NEWS.md
@@ -1,3 +1,10 @@
+# Version 1.7.0
+
+* CIViC update (20240807)
+* NCI Thesaurus update (24.07e)
+* MitelmanDB update (20240715)
+* New dataset: DepMap (cell line) RNA fusion events
+
 # Version 1.6.10 
 
 * Fixed some erroneous drug classifications
@@ -6,24 +13,24 @@
 
 # Version 1.6.8 (June 7th 2024)
 
-* NCI update (24.05d)
+* NCI Thesaurus update (24.05d)
 
 # Version 1.6.7 (May 23rd 2024)
 
-* NCI update (24.04e)
-* Updated CIViC (20240523)
+* NCI Thesaurus update (24.04e)
+* CIViC update (20240523)
 
 # Version 1.6.4 (April 30th 2024)
 
 * Improved clinical (tumor site) annotations of fusions from MitelmanDB
 
 # Version 1.6.3 (April 26th 2024)
 
-* Updated CIViC (20240426)
+* CIViC update (20240426)
 
 # Version 1.6.2 (April 12th 2024)
 
-* Updated NCI Thesaurus (24.03d)
+* NCI Thesaurus update (24.03d)
 
 # Version 1.6.1 (March 26th 2024)
 
@@ -39,7 +46,7 @@
 
 # Version 1.5.8 (February 6th 2024)
 
-* NCI Thesaurus 24.01e
+* NCI Thesaurus update (24.01e)
 
 # Version 1.5.7 (February 3rd 2024)
 
diff --git a/R/sysdata.rda b/R/sysdata.rda
diff --git a/README.md b/README.md
@@ -2,7 +2,7 @@
 
 # pharmOncoX <a href="https://sigven.github.io/pharmOncoX/"><img src="man/figures/logo.png" align="right" height="104" width="90"/></a>
 
-**pharmOncoX** is an R package that provides access to targeted and non-targeted cancer drugs, and genomic cancer biomarkers. Cancer drugs include comprehensive annotations per target, drug mechanism-of-action, approval dates, clinical trial phases for various indications etc. Drugs are further classified according to the [Anatomical Therapeutic Chemical (ATC) Classification System](https://www.whocc.no/atc_ddd_index/), enabling a filtering of cancer drugs according to their main types of action.
+**pharmOncoX** is an R package that provides access to targeted and non-targeted cancer drugs, and genomic cancer biomarkers. Cancer drugs include comprehensive annotations per target, drug mechanism-of-action, approval dates, clinical trial phases for various indications etc. Drugs are further classified according to the [Anatomical Therapeutic Chemical (ATC) Classification System](https://www.whocc.no/atc_ddd_index/), enabling a filtering of cancer drugs according to their main types of action. The package also provides access to data on actionable genomic aberrations (i.e. molecular biomarkers), including gene fusions, mutations, copy number alterations, and expression biomarkers.
 
 
 ## Getting started
diff --git a/data-raw/biomarker_utilities.R b/data-raw/biomarker_utilities.R
@@ -430,7 +430,7 @@ expand_hgvs_terms <- function(var, aa_dict, add_codon_markers = FALSE) {
 }
 
 load_civic_biomarkers <- function(
-    datestamp = '20240130',
+    datestamp = '20240709',
     compound_synonyms = NULL,
     hg38_fasta = 
       "/Users/sigven/research/DB/hg38/hg38.fa",
@@ -2630,8 +2630,26 @@ load_custom_fusion_db <- function() {
   return(biomarker_items)
 }
 
+load_depmap_fusions <- function(db_datestamp = "24Q2"){
+  
+  # Load DepMap fusions
+  depmap_data <- list()
+  depmap_data[['fusions']] <- as.data.frame(read.csv(
+    file = "data-raw/depmap/OmicsFusionFiltered.csv", header = T))
+  
+  depmap_data[['models']] <- as.data.frame(read.csv(
+    file = "data-raw/depmap/Model.csv", header = T)) |>
+    dplyr::select(
+      ModelID, CellLineName, OncotreeLineage,
+      OncotreePrimaryDisease, OncotreeCode,
+      Age, Sex, PrimaryOrMetastasis, SampleCollectionSite,
+      SourceType
+    )
+  return(depmap_data)
+}
+
 load_mitelman_db <- function(cache_dir = NA,
-                             db_datestamp = "20240415") {
+                             db_datestamp = "20240715") {
 
   # Load Mitelman database
   # dos2unix -q -n MBCA.TXT.DATA MBCA.TXT
@@ -2672,7 +2690,7 @@ load_mitelman_db <- function(cache_dir = NA,
 
   fusion_event_data <- as.data.frame(readr::read_tsv(
     file = file.path(
-      cache_dir, "mitelmandb", "MBCA.TXT"),
+      cache_dir, "mitelmandb", "MBCA.TXT.DATA"),
     show_col_types = F, guess_max = 100000)) |>
     dplyr::filter(stringr::str_detect(GeneShort,"::")) |>
     dplyr::rename(variant = GeneShort,
diff --git a/data-raw/custom_drug_target_regex_nci.tsv b/data-raw/custom_drug_target_regex_nci.tsv
@@ -1,6 +1,6 @@
 pattern	symbol
 OMX-0407	SIK3
-RMC-9085	KRAS
+RMC-9085|Olomorasib	KRAS
 3706674	KRAS
 Rineterkib	ERK1
 Rineterkib	ERK2
diff --git a/data-raw/data-raw.R b/data-raw/data-raw.R
@@ -21,7 +21,7 @@ opentargets_version <-
   metadata$compounds[metadata$compounds$source_abbreviation == "opentargets", 
                      "source_version"]
 package_datestamp <- stringr::str_replace_all(Sys.Date(),"-","")
-chembl_pubchem_datestamp <- '20240429' 
+chembl_pubchem_datestamp <- '20240708' 
 
 ## set logging layout
 lgr::lgr$appenders$console$set_layout(
@@ -74,7 +74,7 @@ drug_sets <- list()
 ## Get all anticancer drugs, NCI thesaurus + DGIdb
 drug_sets[['nci']] <- get_nci_drugs(
   nci_db_release = nci_db_release,
-  overwrite = F,
+  overwrite = T,
   path_data_raw = path_data_raw,
   path_data_processed = path_data_tmp_processed)
 
@@ -188,6 +188,8 @@ raw_biomarkers[['mitelmandb']] <-
     cache_dir = file.path(path_data_raw, "biomarkers"))
 raw_biomarkers[['custom_fusions']] <- 
   load_custom_fusion_db()
+raw_biomarkers[['depmap']] <- 
+  load_depmap_fusions()
 
 raw_biomarkers[['custom_fusions']]$variant <- 
   raw_biomarkers[['custom_fusions']]$variant |>
diff --git a/data-raw/drug_name_black_list.txt b/data-raw/drug_name_black_list.txt
@@ -4,6 +4,8 @@ Butanilicaine Hydrochloride
 Tesmilifene Hydrochloride
 8H9 131I
 ABC-294640
+Neladenoson Bialanate
+Sodium Caseinate
 AXL-1717
 AZD-7451
 TAS-115
diff --git a/data-raw/drug_utilities.R b/data-raw/drug_utilities.R
@@ -1750,8 +1750,10 @@ map_curated_targets <- function(gene_info = NULL,
         hit$drug_approved_noncancer <- FALSE
 
         ## set general indications for unknown cases
-        if(is.na(hit$disease_efo_id) & is.na(hit$disease_efo_label) &
-           is.na(hit$cui) & is.na(hit$cui_name)){
+        if(is.na(hit$disease_efo_id) & 
+           is.na(hit$disease_efo_label) &
+           is.na(hit$cui) & 
+           is.na(hit$cui_name)){
           hit$disease_efo_id = "EFO:0000311"
           hit$disease_efo_label = "cancer"
           hit$cui = "C0006826"
diff --git a/data-raw/metadata_pharm_oncox.xlsx b/data-raw/metadata_pharm_oncox.xlsx
diff --git a/pkgdown/index.md b/pkgdown/index.md
@@ -2,20 +2,22 @@
 
 # pharmOncoX <a href="https://sigven.github.io/pharmOncoX/"><img src="man/figures/logo.png" align="right" height="104" width="90"/></a>
 
-**pharmOncoX** provides access to targeted and non-targeted cancer drugs, including comprehensive annotations per target, drug mechanism-of-action, approval dates, clinical trial phases for various indications etc. 
+**pharmOncoX** provides access to targeted and non-targeted cancer drugs, including comprehensive annotations per target, drug mechanism-of-action, approval dates, clinical trial phases for various indications etc. It also provides access to data on actionable genomic aberrations (i.e. molecular biomarkers), including gene fusions, mutations, copy number alterations, and expression biomarkers.
 
 The data is largely based on drug-target-indication associations provided by the [Open Targets Platform](https://targetvalidation.org) ([Ochoa et al., Nucleic Acids Res., 2021](https://doi.org/10.1093/nar/gkaa1027)). Associations retrieved from Open Targets Platform are integrated with cancer-relevant indications/conditions (as provided in [sigven/phenOncoX](https://github.com/sigven/phenOncoX)), allowing the user to retrieve drugs indicated for main tumor types (e.g. `Lung`, `Colon/Rectum` etc.) 
 
 Drug-target associations from the Open Targets Platform have furthermore been integrated and appended with drug information from [NCI Thesaurus](https://ncithesaurus.nci.nih.gov/ncitbrowser/), showing also non-targeted cancer drugs (chemotherapeutic agents etc.), and various drug regimens.
 
 _pharmOncoX_ provides anti-cancer drug classification through existing entries in the [Anatomical Therapeutic Chemical (ATC) Classification System](https://www.whocc.no/atc_ddd_index/), and these have been extended significantly with manual curation, also by establishing novel drug categories that are presently missing in the ATC classificiation tree (examples include _AURK inhibitors_, _MET inhibitors_, _BET inhibitors_, _AKT inhibitors_, _PLK inhibitors_, _IAP inhibitors_, _RAS inhibitors_, _BCL2 inhibitors_ etc.) enabling a filtering of drugs according to their main mechanisms of action.
 
-Currently (as of early June 2024), `pharmOncoX` is built upon the following 
+Currently (as of early August 2024), `pharmOncoX` is built upon the following 
 releases of external databases:
 
- - Open Targets Platform (2024.03)
+ - Open Targets Platform (2024.06)
  - ChEMBL (v34)
- - NCI Thesaurus (24.05d)
+ - NCI Thesaurus (24.07e)
+ - MitelmanDB (20240715)
+ - CIViC (20240807)
 
 ### Getting started
 
diff --git a/vignettes/pharmOncoX.Rmd b/vignettes/pharmOncoX.Rmd
@@ -28,7 +28,7 @@ knitr::opts_chunk$set(
 ```{r install, echo = T, eval = T}
 
 if (!("remotes" %in% installed.packages())) {
- install.packages("remotes")
+  install.packages("remotes")
 }
 
 remotes::install_github('sigven/pharmOncoX')
@@ -190,7 +190,11 @@ drugs <- drugs |>
   dplyr::mutate(
     disease_indication = stringr::str_replace_all(
       disease_indication, "\\|",", ")
-    )
+    ) |>
+  dplyr::select(
+    drug_id, drug_name, drug_type, molecule_chembl_id,
+    drug_action_type, target_symbol, dplyr::everything()
+  )
 
 dt_drugtable_ras_inhibitors <- DT::datatable(
   drugs,
@@ -472,7 +476,11 @@ drugs$records <- drugs$records |>
   dplyr::mutate(
     disease_indication = stringr::str_replace_all(
       disease_indication, "\\|",", ")
-    )
+    ) |>
+  dplyr::select(
+    drug_id, drug_name, drug_type, molecule_chembl_id,
+    drug_action_type, opentargets, dplyr::everything()
+  )
 
 dt_drugtable_platins <- DT::datatable(
   drugs$records,
