use txdbmaker package for makeTxDb... functions (#119)

* use txdbmaker package for makeTxDb... functions * use devel containers for gh actions
rnabioco · Mar 13, 2024 · ee5157e · ee5157e
1 parent 4dbccb8
commit ee5157e
Show file tree

Hide file tree

Showing 7 changed files with 148 additions and 138 deletions.
diff --git a/.github/workflows/check-bioc.yml b/.github/workflows/check-bioc.yml
@@ -53,8 +53,8 @@ jobs:
       fail-fast: false
       matrix:
         config:
-          - { os: ubuntu-latest, r: '4.3', bioc: '3.18', cont: "bioconductor/bioconductor_docker:RELEASE_3_18", rspm: "https://packagemanager.posit.co/cran/__linux__/jammy/latest" }
-          - { os: windows-latest, r: '4.3', bioc: '3.18'}
+          - { os: ubuntu-latest, r: '4.4', bioc: 'devel', cont: "bioconductor/bioconductor_docker:devel", rspm: "https://packagemanager.posit.co/cran/__linux__/jammy/latest" }
+          - { os: windows-latest, r: '4.4', bioc: 'devel'}
           ## Check https://github.com/r-lib/actions/tree/master/examples
           ## for examples using the http-user-agent
     env:
@@ -106,16 +106,16 @@ jobs:
         uses: actions/cache@v3
         with:
           path: ${{ env.R_LIBS_USER }}
-          key: ${{ env.cache-version }}-${{ runner.os }}-biocversion-RELEASE_3_18-r-4.3-${{ hashFiles('.github/depends.Rds') }}
-          restore-keys: ${{ env.cache-version }}-${{ runner.os }}-biocversion-RELEASE_3_18-r-4.3-
+          key: ${{ env.cache-version }}-${{ runner.os }}-biocversion-devel-r-4.4-${{ hashFiles('.github/depends.Rds') }}
+          restore-keys: ${{ env.cache-version }}-${{ runner.os }}-biocversion-devel-r-4.4-
 
       - name: Cache R packages on Linux
         if: "!contains(github.event.head_commit.message, '/nocache') && runner.os == 'Linux' "
         uses: actions/cache@v3
         with:
           path: /home/runner/work/_temp/Library
-          key: ${{ env.cache-version }}-${{ runner.os }}-biocversion-RELEASE_3_18-r-4.3-${{ hashFiles('.github/depends.Rds') }}
-          restore-keys: ${{ env.cache-version }}-${{ runner.os }}-biocversion-RELEASE_3_18-r-4.3-
+          key: ${{ env.cache-version }}-${{ runner.os }}-biocversion-devel-r-4.4-${{ hashFiles('.github/depends.Rds') }}
+          restore-keys: ${{ env.cache-version }}-${{ runner.os }}-biocversion-devel-r-4.4-
 
       - name: Install Linux system dependencies
         if: runner.os == 'Linux'
@@ -289,7 +289,7 @@ jobs:
         if: failure()
         uses: actions/upload-artifact@master
         with:
-          name: ${{ runner.os }}-biocversion-RELEASE_3_18-r-4.3-results
+          name: ${{ runner.os }}-biocversion-devel-r-4.4-results
           path: check
 
         ## Note that DOCKER_PASSWORD is really a token for your dockerhub

diff --git a/DESCRIPTION b/DESCRIPTION
@@ -1,7 +1,7 @@
 Package: raer
 Type: Package
 Title: RNA editing tools in R
-Version: 1.1.4
+Version: 1.1.5
 Authors@R: c(
     person("Kent", "Riemondy", , "kent.riemondy@gmail.com", role = c("aut", "cre"),
            comment = c(ORCID = "0000-0003-0750-1273")),
@@ -55,7 +55,8 @@ Suggests:
     scuttle,
     AnnotationHub,
     covr,
-    raerdata
+    raerdata,
+    txdbmaker 
 LinkingTo: 
     Rhtslib
 SystemRequirements:

diff --git a/R/filter_rse.R b/R/filter_rse.R
@@ -106,25 +106,27 @@ get_splice_sites <- function(txdb, slop = 4) {
 #' @family se-filters
 #'
 #' @examples
-#' library(GenomicFeatures)
-#' rse_adar_ifn <- mock_rse()
-#'
-#' # mock up a txdb with genes
-#' gr <- GRanges(c(
-#'     "DHFR:310-330:-",
-#'     "DHFR:410-415:-",
-#'     "SSR3:100-155:-",
-#'     "SSR3:180-190:-"
-#' ))
-#' gr$source <- "raer"
-#' gr$type <- "exon"
-#' gr$source <- NA
-#' gr$phase <- NA_integer_
-#' gr$gene_id <- c(1, 1, 2, 2)
-#' gr$transcript_id <- rep(c("1.1", "2.1"), each = 2)
-#' txdb <- makeTxDbFromGRanges(gr)
-#'
-#' filter_splice_variants(rse_adar_ifn, txdb)
+#' if(require("txdbmaker")) {
+#'   rse_adar_ifn <- mock_rse()
+#'  
+#'   # mock up a txdb with genes
+#'   gr <- GRanges(c(
+#'       "DHFR:310-330:-",
+#'       "DHFR:410-415:-",
+#'       "SSR3:100-155:-",
+#'       "SSR3:180-190:-"
+#'   ))
+#'   gr$source <- "raer"
+#'   gr$type <- "exon"
+#'   gr$source <- NA
+#'   gr$phase <- NA_integer_
+#'   gr$gene_id <- c(1, 1, 2, 2)
+#'   gr$transcript_id <- rep(c("1.1", "2.1"), each = 2)
+#'   txdb <- txdbmaker::makeTxDbFromGRanges(gr)
+#'  
+#'   filter_splice_variants(rse_adar_ifn, txdb)
+#' }
+#' 
 #'
 #' @returns `SummarizedExperiment::SummarizedExperiment` with sites
 #' adjacent to splice sites removed.
@@ -183,27 +185,28 @@ filter_splice_variants <- function(rse, txdb,
 #'   variants
 #'
 #' @examples
-#' library(GenomicFeatures)
-#'
-#' rse_adar_ifn <- mock_rse()
-#' rse <- rse_adar_ifn[seqnames(rse_adar_ifn) == "SPCS3"]
-#'
-#' # mock up a txdb with genes
-#' gr <- GRanges(c(
-#'     "SPCS3:100-120:-",
-#'     "SPCS3:325-350:-"
-#' ))
-#' gr$source <- "raer"
-#' gr$type <- "exon"
-#' gr$source <- NA
-#' gr$phase <- NA_integer_
-#' gr$gene_id <- c(1, 2)
-#' gr$transcript_id <- c("1.1", "2.1")
-#' txdb <- makeTxDbFromGRanges(gr)
-#'
-#' rse <- filter_multiallelic(rse)
-#' filter_clustered_variants(rse, txdb, variant_dist = 10)
+#' if(require("txdbmaker")){
+#'   rse_adar_ifn <- mock_rse()
+#'   rse <- rse_adar_ifn[seqnames(rse_adar_ifn) == "SPCS3"]
+#'  
+#'   # mock up a txdb with genes
+#'   gr <- GRanges(c(
+#'       "SPCS3:100-120:-",
+#'       "SPCS3:325-350:-"
+#'   ))
+#'   gr$source <- "raer"
+#'   gr$type <- "exon"
+#'   gr$source <- NA
+#'   gr$phase <- NA_integer_
+#'   gr$gene_id <- c(1, 2)
+#'   gr$transcript_id <- c("1.1", "2.1")
+#'   txdb <- txdbmaker::makeTxDbFromGRanges(gr)
+#'  
+#'   rse <- filter_multiallelic(rse)
+#'   filter_clustered_variants(rse, txdb, variant_dist = 10)
 #'
+#' }
+#' 
 #' @family se-filters
 #'
 #' @return `SummarizedExperiment::SummarizedExperiment` with sites removed from

diff --git a/man/filter_clustered_variants.Rd b/man/filter_clustered_variants.Rd
diff --git a/man/filter_splice_variants.Rd b/man/filter_splice_variants.Rd
diff --git a/tests/testthat/test_rse.R b/tests/testthat/test_rse.R
@@ -1,6 +1,5 @@
 pkgs <- c(
-    "GenomicRanges", "GenomicFeatures",
-    "SummarizedExperiment", "rtracklayer"
+    "GenomicRanges", "SummarizedExperiment", "rtracklayer", "GenomicFeatures"
 )
 
 msg <- lapply(pkgs, function(x) {
@@ -16,7 +15,7 @@ sites <- import(bedfn)
 res <- pileup_sites(bamfn, fafn)
 
 test_that("annot_snps works", {
-    if (require(SNPlocs.Hsapiens.dbSNP144.GRCh38)) {
+    if (require("SNPlocs.Hsapiens.dbSNP144.GRCh38")) {
         gr <- GRanges(rep("22", 10),
             IRanges(
                 seq(10510077,
@@ -43,7 +42,7 @@ test_that("annot_snps works", {
         expect_true(is(res, "RangedSummarizedExperiment"))
         expect_true("RefSNP_id" %in% colnames(mcols(res)))
 
-        if (require(BSgenome.Hsapiens.NCBI.GRCh38)) {
+        if (require("BSgenome.Hsapiens.NCBI.GRCh38")) {
             res <- annot_snps(se,
                 dbsnp = SNPlocs.Hsapiens.dbSNP144.GRCh38,
                 genome = BSgenome.Hsapiens.NCBI.GRCh38
@@ -121,62 +120,66 @@ mock_txdb <- function() {
     gr$phase <- NA_integer_
     gr$gene_id <- c(1, 1, 2, 2)
     gr$transcript_id <- rep(c("1.1", "2.1"), each = 2)
-    makeTxDbFromGRanges(gr)
+    txdbmaker::makeTxDbFromGRanges(gr)
 }
 
 test_that("get_splice_sites and filtering works", {
-    txdb <- mock_txdb()
-    spl_sites <- get_splice_sites(txdb,
-        slop = 3
-    )
-    expect_true(all(width(spl_sites) == 6))
-    expect_error(get_splice_sites(spl_sites))
-
-    rse_adar_ifn <- mock_rse()
-    expect_message(rse <- filter_splice_variants(rse_adar_ifn, txdb))
-    n_removed <- nrow(subsetByOverlaps(rse_adar_ifn, rse, invert = TRUE))
-    expect_equal(n_removed, 5L)
-    expect_true("site_DHFR_328_2" %in% rownames(rse_adar_ifn))
-    expect_false("site_DHFR_328_2" %in% rownames(rse))
+    if("txdbmaker" %in% rownames(installed.packages())) {
+        txdb <- mock_txdb()
+        spl_sites <- get_splice_sites(txdb,
+                                      slop = 3
+        )
+        expect_true(all(width(spl_sites) == 6))
+        expect_error(get_splice_sites(spl_sites))
+
+        rse_adar_ifn <- mock_rse()
+        expect_message(rse <- filter_splice_variants(rse_adar_ifn, txdb))
+        n_removed <- nrow(subsetByOverlaps(rse_adar_ifn, rse, invert = TRUE))
+        expect_equal(n_removed, 5L)
+        expect_true("site_DHFR_328_2" %in% rownames(rse_adar_ifn))
+        expect_false("site_DHFR_328_2" %in% rownames(rse))
+    }
 })
 
 test_that("removing clustered variants works", {
-    txdb <- mock_txdb()
-
-    rse_adar_ifn <- mock_rse()
-    expect_message(rse <- filter_multiallelic(rse_adar_ifn))
-
-    gr <- rowRanges(rse)
-    nd <- distanceToNearest(gr)
-    gr1 <- gr[mcols(nd)$distance > 100]
-    expect_warning(expect_warning(gr2 <- filter_clustered_variants(rse, txdb)))
-    expect_true(identical(rowRanges(gr2), gr1))
-
-    # check that transcript based removal works
-    # DHFR_328 -> DHFR_423 overlaps 310-330 -> 410-430 splice
-    ex <- exons(txdb)
-    rse_ex <- subsetByOverlaps(rse, ex)
-
-    expect_warning(
-        gr2 <- filter_clustered_variants(rse_ex,
-            txdb,
-            regions = "transcript",
-            variant_dist = 20
+    if("txdbmaker" %in% rownames(installed.packages())) {
+        txdb <- mock_txdb()
+
+        rse_adar_ifn <- mock_rse()
+        expect_message(rse <- filter_multiallelic(rse_adar_ifn))
+
+        gr <- rowRanges(rse)
+        nd <- distanceToNearest(gr)
+        gr1 <- gr[mcols(nd)$distance > 100]
+        expect_warning(expect_warning(gr2 <- filter_clustered_variants(rse, txdb)))
+        expect_true(identical(rowRanges(gr2), gr1))
+
+        # check that transcript based removal works
+        # DHFR_328 -> DHFR_423 overlaps 310-330 -> 410-430 splice
+        ex <- exons(txdb)
+        rse_ex <- subsetByOverlaps(rse, ex)
+
+        expect_warning(
+            gr2 <- filter_clustered_variants(rse_ex,
+                                             txdb,
+                                             regions = "transcript",
+                                             variant_dist = 20
+            )
         )
-    )
-
-    expt <- rowRanges(subsetByOverlaps(rse_ex, gr2, invert = T))
-    ds <- distanceToNearest(mapToTranscripts(expt, txdb,
-        extractor.fun = exonsBy
-    ))
-    expect_true(all(mcols(ds)$distance < 20))
-
-    expect_warning(
-        gr2 <- filter_clustered_variants(rse_ex, txdb,
-            variant_dist = 25
+        
+        expt <- rowRanges(subsetByOverlaps(rse_ex, gr2, invert = T))
+        ds <- distanceToNearest(mapToTranscripts(expt, txdb,
+                                                 extractor.fun = exonsBy
+        ))
+        expect_true(all(mcols(ds)$distance < 20))
+        
+        expect_warning(
+            gr2 <- filter_clustered_variants(rse_ex, txdb,
+                                             variant_dist = 25
+            )
         )
-    )
-    expect_equal(length(gr2), 3L)
+        expect_equal(length(gr2), 3L)
+    }
 })
 
 test_that("calc_confidence works", {

diff --git a/vignettes/raer.Rmd b/vignettes/raer.Rmd
@@ -576,7 +576,7 @@ region of chr4.
 
 - `TxDb.Hsapiens.UCSC.hg38.knownGene`, a database of transcript models. 
 Alternatively these can be generated from a `.gtf` file using 
-`makeTxDbFromGRanges()` from the `GenomicFeatures` package.   
+`makeTxDbFromGRanges()` from the `txdbmaker` package.   
 
 - RepeatMasker annotations, which can be obtained from the `AnnotationHub()` 
 for hg38, as shown in the [bulk RNA-seq](#bulk) tutorial.