Update latest cosmic database to v3.4

ShixiangWang · Nov 14, 2023 · 3eb636c · 3eb636c
1 parent 35d12c6
commit 3eb636c
Show file tree

Hide file tree

Showing 5 changed files with 32 additions and 16 deletions.
diff --git a/.gitignore b/.gitignore
@@ -21,3 +21,8 @@ inst/extdata/mouse_mm9_gene_info.rds
 /doc/
 /Meta/
 inst/extdata/COSMIC_v3.3_CN_GRCh37.rds
+inst/extdata/COSMIC_v3.4_CN_GRCh37.rds
+inst/extdata/COSMIC_v3.4_DBS_GRCh37.rds
+inst/extdata/COSMIC_v3.4_RNA-SBS_GRCh37.rds
+inst/extdata/COSMIC_v3.4_SBS_GRCh37.rds
+inst/extdata/COSMIC_v3.4_SV_GRCh38.rds
diff --git a/NEWS.md b/NEWS.md
@@ -1,3 +1,12 @@
+# sigminer 2.3.0
+
+- Updated COSMIC database to v3.4. SV and RNA-SBS signatures are included.
+
+```r
+get_sig_db("latest_RNA-SBS_GRCh37")
+get_sig_db("latest_SV_GRCh38")
+```
+
 # sigminer 2.2.2
 
 - Updated pkg doc following the new CRAN feature (thanks to K from the CRAN team).

diff --git a/R/get_sig_db.R b/R/get_sig_db.R
@@ -120,23 +120,25 @@ get_sig_db <- function(sig_db = "legacy") {
           package = "sigminer", mustWork = TRUE
         )
       } else if (startsWith(sig_db, "latest_")) {
-        latest_version <- "v3.3"
+        latest_version <- "v3.4"
         # v3.2_SBS_GRCh37.txt
         sig_db2 <- sub("latest_", "", sig_db)
 
         file_dir <- switch(sig_db,
-          latest_SBS_GRCh37 = "1908",
-          latest_DBS_GRCh37 = "1902",
-          latest_ID_GRCh37 = "1907",
-          latest_SBS_GRCh38 = "1909",
-          latest_DBS_GRCh38 = "1903",
-          latest_SBS_mm9 = "1910",
-          latest_DBS_mm9 = "1904",
-          latest_SBS_mm10 = "1911",
-          latest_DBS_mm10 = "1905",
-          latest_SBS_rn6 = "1912",
-          latest_DBS_rn6 = "1906",
-          latest_CN_GRCh37 = "2044",
+          latest_SBS_GRCh37 = "2123",
+          latest_DBS_GRCh37 = "2116",
+          latest_ID_GRCh37 = "2121",
+          latest_SBS_GRCh38 = "2124",
+          latest_DBS_GRCh38 = "2117",
+          latest_SBS_mm9 = "2125",
+          latest_DBS_mm9 = "2118",
+          latest_SBS_mm10 = "2126",
+          latest_DBS_mm10 = "2119",
+          latest_SBS_rn6 = "2127",
+          latest_DBS_rn6 = "2120",
+          latest_CN_GRCh37 = "2115",
+          `latest_RNA-SBS_GRCh37` = "2122",
+          latest_SV_GRCh38 = "2128",
           stop("Bad input!")
         )
 
@@ -153,7 +155,7 @@ get_sig_db <- function(sig_db = "legacy") {
           message("The data is not available in local, obtain it from COSMIC: https://cancer.sanger.ac.uk/signatures/downloads/")
           data <- tryCatch(
             data.table::fread(data_url, data.table = FALSE) %>%
-              tibble::column_to_rownames("Type") %>%
+              tibble::column_to_rownames(if (sig_db == "latest_RNA-SBS_GRCh37") "V1" else "Type") %>%
               as.matrix(),
             error = function(e) {
               message(e$message)

diff --git a/R/get_sig_similarity.R b/R/get_sig_similarity.R
@@ -114,7 +114,7 @@ get_sig_similarity <- function(Signature, Ref = NULL,
                                  "latest_SBS_mm9", "latest_DBS_mm9",
                                  "latest_SBS_mm10", "latest_DBS_mm10",
                                  "latest_SBS_rn6", "latest_DBS_rn6",
-                                 "latest_CN_GRCh37"
+                                 "latest_CN_GRCh37", "latest_RNA-SBS_GRCh37", "latest_SV_GRCh38"
                                ),
                                db_type = c("", "human-exome", "human-genome"),
                                method = "cosine",

diff --git a/R/sig_fit.R b/R/sig_fit.R
@@ -113,7 +113,7 @@ sig_fit <- function(catalogue_matrix,
                       "latest_SBS_mm9", "latest_DBS_mm9",
                       "latest_SBS_mm10", "latest_DBS_mm10",
                       "latest_SBS_rn6", "latest_DBS_rn6",
-                      "latest_CN_GRCh37"
+                      "latest_CN_GRCh37", "latest_RNA-SBS_GRCh37", "latest_SV_GRCh38"
                     ),
                     db_type = c("", "human-exome", "human-genome"),
                     show_index = TRUE,