rescue parseCSQToGRanges from ensemblVEP

kevinrue · Jul 4, 2024 · 8c109fa · 8c109fa
1 parent ddaaabe
commit 8c109fa
Show file tree

Hide file tree

Showing 7 changed files with 139 additions and 59 deletions.
diff --git a/.github/workflows/check-bioc.yml b/.github/workflows/check-bioc.yml
@@ -52,9 +52,9 @@ jobs:
       fail-fast: false
       matrix:
         config:
-          - { os: ubuntu-latest, r: '4.1', bioc: '3.14', cont: "bioconductor/bioconductor_docker:devel", rspm: "https://packagemanager.rstudio.com/cran/__linux__/focal/latest" }
-          - { os: macOS-latest, r: '4.1', bioc: '3.14'}
-          ##- { os: windows-latest, r: '4.1', bioc: '3.14'}
+          - { os: ubuntu-latest, r: 'devel', bioc: '3.20', cont: "bioconductor/bioconductor_docker:devel", rspm: "https://packagemanager.rstudio.com/cran/__linux__/jammy/latest" }
+          - { os: macOS-latest, r: 'devel', bioc: '3.20'}
+          # - { os: windows-latest, r: 'devel', bioc: '3.20'}
           ## Check https://github.com/r-lib/actions/tree/master/examples
           ## for examples using the http-user-agent
     env:
@@ -79,20 +79,20 @@ jobs:
       ## https://github.com/r-lib/actions/blob/master/examples/check-standard.yaml
       ## If they update their steps, we will also need to update ours.
       - name: Checkout Repository
-        uses: actions/checkout@v2
+        uses: actions/checkout@v3
 
       ## R is already included in the Bioconductor docker images
       - name: Setup R from r-lib
         if: runner.os != 'Linux'
-        uses: r-lib/actions/setup-r@master
+        uses: r-lib/actions/setup-r@v2
         with:
           r-version: ${{ matrix.config.r }}
           http-user-agent: ${{ matrix.config.http-user-agent }}
 
       ## pandoc is already included in the Bioconductor docker images
       - name: Setup pandoc from r-lib
         if: runner.os != 'Linux'
-        uses: r-lib/actions/setup-pandoc@master
+        uses: r-lib/actions/setup-pandoc@v2
 
       - name: Query dependencies
         run: |
@@ -102,19 +102,19 @@ jobs:
 
       - name: Restore R package cache
         if: "!contains(github.event.head_commit.message, '/nocache') && runner.os != 'Linux'"
-        uses: actions/cache@v2
+        uses: actions/cache@v3
         with:
           path: ${{ env.R_LIBS_USER }}
-          key: ${{ env.cache-version }}-${{ runner.os }}-biocversion-devel-r-4.1-${{ hashFiles('.github/depends.Rds') }}
-          restore-keys: ${{ env.cache-version }}-${{ runner.os }}-biocversion-devel-r-4.1-
+          key: ${{ env.cache-version }}-${{ runner.os }}-biocversion-devel-r-devel-${{ hashFiles('.github/depends.Rds') }}
+          restore-keys: ${{ env.cache-version }}-${{ runner.os }}-biocversion-devel-r-devel-
 
       - name: Cache R packages on Linux
         if: "!contains(github.event.head_commit.message, '/nocache') && runner.os == 'Linux' "
-        uses: actions/cache@v2
+        uses: actions/cache@v3
         with:
           path: /home/runner/work/_temp/Library
-          key: ${{ env.cache-version }}-${{ runner.os }}-biocversion-devel-r-4.1-${{ hashFiles('.github/depends.Rds') }}
-          restore-keys: ${{ env.cache-version }}-${{ runner.os }}-biocversion-devel-r-4.1-
+          key: ${{ env.cache-version }}-${{ runner.os }}-biocversion-devel-r-devel-${{ hashFiles('.github/depends.Rds') }}
+          restore-keys: ${{ env.cache-version }}-${{ runner.os }}-biocversion-devel-r-devel-
 
       - name: Install Linux system dependencies
         if: runner.os == 'Linux'
@@ -143,6 +143,9 @@ jobs:
           ## Required for tcltk
           brew install xquartz --cask
 
+          ## Required for terra
+          brew install gdal
+
       - name: Install Windows system dependencies
         if: runner.os == 'Windows'
         run: |
@@ -176,7 +179,7 @@ jobs:
           gha_repos <- if(
               .Platform$OS.type == "unix" && Sys.info()["sysname"] != "Darwin"
           ) c(
-              "AnVIL" = "https://bioconductordocker.blob.core.windows.net/packages/3.14/bioc",
+              "AnVIL" = "https://bioconductordocker.blob.core.windows.net/packages/3.20/bioc",
               BiocManager::repositories()
               ) else BiocManager::repositories()
 
@@ -210,15 +213,15 @@ jobs:
         shell: Rscript {0}
 
       - name: Install covr
-        if: github.ref == 'refs/heads/master' && env.run_covr == 'true' && runner.os == 'Linux'
+        if: github.ref == 'refs/heads/devel' && env.run_covr == 'true' && runner.os == 'Linux'
         run: |
           remotes::install_cran("covr")
         shell: Rscript {0}
 
       - name: Install pkgdown
-        if: github.ref == 'refs/heads/master' && env.run_pkgdown == 'true' && runner.os == 'Linux'
+        if: github.ref == 'refs/heads/devel' && env.run_pkgdown == 'true' && runner.os == 'Linux'
         run: |
-          remotes::install_github("r-lib/pkgdown")
+          remotes::install_cran("pkgdown")
         shell: Rscript {0}
 
       - name: Session info
@@ -259,27 +262,28 @@ jobs:
         run: |
           BiocCheck::BiocCheck(
               dir('check', 'tar.gz$', full.names = TRUE),
-              `quit-with-status` = TRUE,
+              `quit-with-status` = FALSE,
               `no-check-R-ver` = TRUE,
               `no-check-bioc-help` = TRUE
           )
         shell: Rscript {0}
 
       - name: Test coverage
-        if: github.ref == 'refs/heads/master' && env.run_covr == 'true' && runner.os == 'Linux'
+        if: github.ref == 'refs/heads/devel' && env.run_covr == 'true' && runner.os == 'Linux'
         run: |
           covr::codecov()
         shell: Rscript {0}
 
       - name: Install package
-        if: github.ref == 'refs/heads/master' && env.run_pkgdown == 'true' && runner.os == 'Linux'
+        if: github.ref == 'refs/heads/devel' && env.run_pkgdown == 'true' && runner.os == 'Linux'
         run: R CMD INSTALL .
 
       - name: Build and deploy pkgdown site
-        if: github.ref == 'refs/heads/master' && env.run_pkgdown == 'true' && runner.os == 'Linux'
+        if: github.ref == 'refs/heads/devel' && env.run_pkgdown == 'true' && runner.os == 'Linux'
         run: |
-          git config --local user.name "$GITHUB_ACTOR"
-          git config --local user.email "[email protected]"
+          git config --global user.name "$GITHUB_ACTOR"
+          git config --global user.email "[email protected]"
+          git config --global --add safe.directory /__w/kevinrue/TVTB
           Rscript -e "pkgdown::deploy_to_branch(new_process = FALSE)"
         shell: bash {0}
         ## Note that you need to run pkgdown::deploy_to_branch(new_process = FALSE)
@@ -291,7 +295,7 @@ jobs:
         if: failure()
         uses: actions/upload-artifact@master
         with:
-          name: ${{ runner.os }}-biocversion-devel-r-4.1-results
+          name: ${{ runner.os }}-biocversion-devel-r-devel-results
           path: check
 
       - uses: docker/build-push-action@v1

diff --git a/DESCRIPTION b/DESCRIPTION
@@ -1,8 +1,8 @@
 Package: TVTB
 Type: Package
 Title: TVTB: The VCF Tool Box
-Version: 1.31.0
-Date: 2021-08-30
+Version: 1.31.1
+Date: 2024-07-04
 Authors@R: person("Kevin", "Rue-Albrecht", role = c("aut", "cre"),
     email = "[email protected]")
 Description: The package provides S4 classes and methods to filter,
@@ -14,20 +14,47 @@ Description: The package provides S4 classes and methods to filter,
     Shiny Variant Explorer (tSVE).
 License: Artistic-2.0
 Depends: R (>= 3.4), methods, utils, stats
-Imports: AnnotationFilter, BiocGenerics (>= 0.25.1), BiocParallel, Biostrings,
-    ensembldb, ensemblVEP, GenomeInfoDb, GenomicRanges, GGally, ggplot2, Gviz,
-    limma, IRanges (>= 2.21.6), reshape2, Rsamtools, S4Vectors (>= 0.25.14),
-    SummarizedExperiment, VariantAnnotation (>= 1.19.9)
+Imports:
+    AnnotationFilter,
+    BiocGenerics (>= 0.25.1),
+    BiocParallel,
+    Biostrings,
+    ensembldb,
+    GenomeInfoDb,
+    GenomicRanges,
+    GGally,
+    ggplot2,
+    Gviz,
+    limma,
+    IRanges (>= 2.21.6),
+    reshape2,
+    Rsamtools,
+    S4Vectors (>= 0.25.14),
+    SummarizedExperiment,
+    VariantAnnotation (>= 1.19.9)
 Suggests: EnsDb.Hsapiens.v75 (>= 0.99.7), shiny (>= 0.13.2.9005), DT
     (>= 0.1.67), rtracklayer, BiocStyle (>= 2.5.19), knitr (>= 1.12),
     rmarkdown, testthat, covr, pander
-biocViews: Software, Genetics, GeneticVariability, GenomicVariation,
-    DataRepresentation, GUI, Genetics, DNASeq, WholeGenome,
-    Visualization, MultipleComparison, DataImport,
-    VariantAnnotation, Sequencing, Coverage, Alignment,
+biocViews:
+    Software,
+    Genetics,
+    GeneticVariability,
+    GenomicVariation,
+    DataRepresentation,
+    GUI,
+    Genetics,
+    DNASeq,
+    WholeGenome,
+    Visualization,
+    MultipleComparison,
+    DataImport,
+    VariantAnnotation,
+    Sequencing,
+    Coverage,
+    Alignment,
     SequenceMatching
 Collate: utils.R tSVE.R AllClasses.R AllGenerics.R Genotypes-class.R
-    TVTBparam-class.R VcfFilterRules-class.R countGenos-methods.R
+    TVTBparam-class.R VcfFilterRules-class.R parseCSQToGRanges.R countGenos-methods.R
     autodetectGenotypes.R addCountGenos-methods.R addFrequencies-methods.R
     addOverallFrequencies-methods.R
     addPhenoLevelFrequencies-methods.R dropInfo.R readVcf-methods.R
@@ -36,4 +63,3 @@ Collate: utils.R tSVE.R AllClasses.R AllGenerics.R Genotypes-class.R
 VignetteBuilder: knitr
 URL: https://github.com/kevinrue/TVTB
 BugReports: https://github.com/kevinrue/TVTB/issues
-RoxygenNote: 7.1.1
diff --git a/NAMESPACE b/NAMESPACE
@@ -71,8 +71,6 @@ importFrom(
     "vcfSamples<-", "vcfWhich<-", "vcfFixed<-", "vcfInfo<-", "vcfGeno<-"
 )
 
-importFrom("ensemblVEP", "parseCSQToGRanges")
-
 importFrom(
     "ggplot2",
     "ggplot", "aes_string", "facet_wrap",
@@ -96,6 +94,12 @@ importFrom(
     "reshape2",
     "melt", "dcast")
 
+importFrom(
+    "GenomeInfoDb",
+    "genome",
+    "genome<-"
+)
+
 #exportPattern("^[[:alpha:]]+")
 
 # runShinyApp.R ----

diff --git a/R/AllGenerics.R b/R/AllGenerics.R
@@ -187,3 +187,10 @@ setGeneric(
     function(x, value)
         standardGeneric("svp<-")
 )
+
+# ensemblVEP ----
+
+setGeneric("parseCSQToGRanges", signature = "x",
+  function(x, VCFRowID=TRUE, ...)
+    standardGeneric("parseCSQToGRanges")
+)
diff --git a/R/parseCSQToGRanges.R b/R/parseCSQToGRanges.R
@@ -0,0 +1,41 @@
+setMethod("parseCSQToGRanges", "VCF",
+  function(x, VCFRowID=character(), ..., info.key="CSQ")
+  {
+    ## no 'info.key'
+    if (!info.key %in% names(info(x)))
+      return(rowRanges(x))
+
+    hdr <- info(header(x))[info.key, "Description"]
+    nms <- unlist(strsplit(strsplit(hdr, "Format: ")[[1]][2], "\\|"))
+    ulst <- unlist(info(x)[[info.key]], use.names=FALSE)
+    ## 'info.key' without data
+    if (all(is.na(ulst))) {
+      gr <- rowRanges(x)
+      csq <-
+        DataFrame(setNames(replicate(length(nms), character(0)), nms))
+    } else {
+      ## 'info.key' with data
+      elt <- elementNROWS(info(x)[[info.key]])
+      raw <- strsplit(ulst, "\\|")
+      csq <- matrix(nrow=length(ulst), ncol=length(nms))
+      for (i in 1:nrow(csq))
+        csq[i, 1:length(raw[[i]])] <- raw[[i]]
+      csq[!nzchar(csq)] <- NA
+      colnames(csq) <- nms
+      csq <- data.frame(csq, stringsAsFactors=FALSE)
+
+      rd <- rowRanges(x)
+      gr <- rd[rep(seq_along(rd), elt)]
+      if (length(VCFRowID)) {
+        if (any(no_match <- !VCFRowID %in% rownames(x)))
+          warning(paste0("rownames not found in 'x' : ",
+            paste(VCFRowID[no_match], collapse=",")))
+        VCFRowID <- rep(match(rownames(x), VCFRowID), elt)
+        csq <- DataFrame(VCFRowID=VCFRowID, csq)
+      }
+    }
+    mcols(gr) <- csq
+    genome(gr) <- genome(x)
+    gr
+  }
+)
diff --git a/inst/NEWS.Rd b/inst/NEWS.Rd
@@ -1,6 +1,14 @@
 \name{TVTB-NEWS}
 \title{NEWS: The VCF Tool Box}
 
+\section{Changes in version 1.31.1 2024-07-04}{
+\subsection{Bug fix}{
+\itemize{
+\item Copy \code{parseCSQToGRanges} from \code{ensemblVEP}
+}
+}
+}
+
 \section{Changes in version 1.19.1 2021-08-30}{
 \subsection{Bug fix}{
 \itemize{

diff --git a/man/VcfBasicRules-class.Rd b/man/VcfBasicRules-class.Rd
@@ -109,7 +109,6 @@ decribed in this help page, except when specified otherwise.
 }
 
 \section{Constructors}{
-\describe{}{
 \code{VcfFixedRules(exprs = list(), ..., active = TRUE)}
 
 \code{VcfInfoRules(exprs = list(), ..., active = TRUE)}
@@ -123,32 +122,23 @@ which is recycled as necessary.
 
 See the constructor of \code{FilterRules} for more details.
 }
-}
 
 \section{Subsetting and Replacement}{
 In the following code snippets \code{x} and \code{value}
 are objects from any of the classes described in this help page.
 
-\describe{
-    \item{}{
-    \code{x[i]}: Subsets the filter rules using the same interface as for
-    \code{\linkS4class{List}}.
-    }
-    \item{}{
-    \code{x[[i]]}: Extracts an expression or function via the same interface
-    as for \code{\linkS4class{List}}.
-    }
-    \item{}{
-    \code{x[i] <- value}: Replaces a filter rule by one of the
-    \strong{same} class.
-    The active state(s) and name(s) are transferred from \code{value} to
-    \code{x}.
-    }
-    \item{}{
-    \code{x[[i]] <- value}:
-    The same interface as for \code{\linkS4class{List}}.
-    The default active state for new rules is TRUE.
-    }
+\itemize{
+\item{\code{x[i]}: Subsets the filter rules using the same interface as for
+\code{\linkS4class{List}}.}
+\item{\code{x[[i]]}: Extracts an expression or function via the same interface
+as for \code{\linkS4class{List}}.}
+\item{\code{x[i] <- value}: Replaces a filter rule by one of the
+\strong{same} class.
+The active state(s) and name(s) are transferred from \code{value} to
+\code{x}.}
+\item{\code{x[[i]] <- value}:
+The same interface as for \code{\linkS4class{List}}.
+The default active state for new rules is TRUE.}
 }
 }