Merge branch 'develop'

stuart-lab · Dec 8, 2021 · 63d61be · 63d61be
2 parents 10d950e + 2bf638b
commit 63d61be
Show file tree

Hide file tree

Showing 27 changed files with 263 additions and 86 deletions.
diff --git a/.github/workflows/cmd-check.yml b/.github/workflows/cmd-check.yml
@@ -90,24 +90,24 @@ jobs:
       - name: Query dependencies
         run: |
           install.packages('remotes')
-          saveRDS(remotes::dev_package_deps(dependencies = TRUE), ".github/depends.Rds", version = 2)
+          saveRDS(remotes::dev_package_deps(), ".github/depends.Rds", version = 2)
         shell: Rscript {0}
 
       - name: Cache R packages
         if: "!contains(github.event.head_commit.message, '/nocache') && runner.os != 'Linux'"
         uses: actions/cache@v2
         with:
           path: ${{ env.R_LIBS_USER }}
-          key: ${{ env.cache-version }}-${{ runner.os }}-biocversion-RELEASE_3_12-r-4.0-${{ hashFiles('.github/depends.Rds') }}
-          restore-keys: ${{ env.cache-version }}-${{ runner.os }}-biocversion-RELEASE_3_12-r-4.0-
+          key: ${{ env.cache-version }}-${{ runner.os }}-biocversion-RELEASE_3_12-r-4.0-2${{ hashFiles('.github/depends.Rds') }}
+          restore-keys: ${{ env.cache-version }}-${{ runner.os }}-biocversion-RELEASE_3_12-r-4.0-2
 
       - name: Cache R packages on Linux
         if: "!contains(github.event.head_commit.message, '/nocache') && runner.os == 'Linux' "
         uses: actions/cache@v2
         with:
           path: /home/runner/work/_temp/Library
-          key: ${{ env.cache-version }}-${{ runner.os }}-biocversion-RELEASE_3_12-r-4.0-${{ hashFiles('.github/depends.Rds') }}
-          restore-keys: ${{ env.cache-version }}-${{ runner.os }}-biocversion-RELEASE_3_12-r-4.0-
+          key: ${{ env.cache-version }}-${{ runner.os }}-biocversion-RELEASE_3_12-r-4.0-2${{ hashFiles('.github/depends.Rds') }}
+          restore-keys: ${{ env.cache-version }}-${{ runner.os }}-biocversion-RELEASE_3_12-r-4.0-2
 
       - name: Install Linux system dependencies
         if: runner.os == 'Linux'
@@ -161,15 +161,15 @@ jobs:
 
           ## Pass #1 at installing dependencies
           message(paste('****', Sys.time(), 'pass number 1 at installing dependencies: local dependencies ****'))
-          remotes::install_local(dependencies = TRUE, repos = BiocManager::repositories(), build_vignettes = TRUE, upgrade = TRUE)
+          remotes::install_local(repos = BiocManager::repositories(), build_vignettes = TRUE, upgrade = TRUE)
         continue-on-error: true
         shell: Rscript {0}
 
       - name: Install dependencies pass 2
         run: |
           ## Pass #2 at installing dependencies
           message(paste('****', Sys.time(), 'pass number 2 at installing dependencies: any remaining dependencies ****'))
-          remotes::install_local(dependencies = TRUE, repos = BiocManager::repositories(), build_vignettes = TRUE, upgrade = TRUE)
+          remotes::install_local(repos = BiocManager::repositories(), build_vignettes = TRUE, upgrade = TRUE)
 
           ## For running the checks
           message(paste('****', Sys.time(), 'installing rcmdcheck and BiocCheck ****'))
@@ -194,6 +194,8 @@ jobs:
       - name: Run CMD check
         env:
           _R_CHECK_CRAN_INCOMING_: false
+          _R_CHECK_DEPENDS_ONLY_: true
+          _R_CHECK_FORCE_SUGGESTS_: false
         run: |
           rcmdcheck::rcmdcheck(
               args = c("--no-build-vignettes", "--no-manual", "--timings"),

diff --git a/DESCRIPTION b/DESCRIPTION
@@ -1,7 +1,7 @@
 Package: Signac
 Title: Analysis of Single-Cell Chromatin Data
-Version: 1.4.0
-Date: 2021-09-20
+Version: 1.5.0
+Date: 2021-12-07
 Authors@R: c(
   person(given = 'Tim', family = 'Stuart', email = 'tstuart@nygenome.org', role = c('aut', 'cre'), comment = c(ORCID = '0000-0002-3044-0897')),
   person(given = 'Avi', family = 'Srivastava', email = 'asrivastava@nygenome.org', role = 'aut', comment = c(ORCID = '0000-0001-9798-2079')),
@@ -12,14 +12,14 @@ Description: A framework for the analysis and exploration of single-cell chromat
     The 'Signac' package contains functions for quantifying single-cell chromatin data,
     computing per-cell quality control metrics, dimension reduction
     and normalization, visualization, and DNA sequence motif analysis.
-    Reference: Stuart et al. (2020) <doi:10.1101/2020.11.09.373613>.
+    Reference: Stuart et al. (2021) <doi:10.1038/s41592-021-01282-5>.
 Depends:
     R (>= 4.0.0),
     methods
 License: MIT + file LICENSE
 Encoding: UTF-8
 LazyData: true
-RoxygenNote: 7.1.1
+RoxygenNote: 7.1.2
 URL: https://github.com/timoast/signac, https://satijalab.org/signac
 BugReports: https://github.com/timoast/signac/issues
 LinkingTo: Rcpp
@@ -75,6 +75,7 @@ Collate:
     'quantification.R'
     'utilities.R'
     'visualization.R'
+    'zzz.R'
 Suggests: 
     testthat (>= 2.1.0),
     chromVAR,

diff --git a/NEWS.md b/NEWS.md
@@ -1,3 +1,20 @@
+# Signac 1.5.0
+
+Bug fixes:
+
+* Fixed bug in `FeatureMatrix()` when cells information not present in Fragment object ([#803](https://github.com/timoast/signac/issues/803))
+* Fixed bug in object merging ([#804](https://github.com/timoast/signac/issues/804))
+* Add ability to run `LinkPeaks()` using Ensembl IDs ([#858](https://github.com/timoast/signac/issues/858))
+* Fix issue in `GeneActivity()` when gene names are `NA` ([#865](https://github.com/timoast/signac/issues/865))
+* Fix bug in `FeatureMatrix()` when only one region supplied
+* Allow negative values in `ExpressionPlot()` when using scaled data ([#893](https://github.com/timoast/signac/issues/893))
+
+Other changes:
+
+* Added `idf` parameter to `RunTFIDF()` to use precomputed IDF vector
+* Added `gene.id` parameter to `GeneActivity()` to allow output genes named using gene ID ([#837](https://github.com/timoast/signac/issues/837))
+* Added `sep` parameter to `ConnectionsToLinks()` ([#841](https://github.com/timoast/signac/issues/841))
+
 # Signac 1.4.0
 
 Bug fixes:

diff --git a/R/links.R b/R/links.R
@@ -86,6 +86,9 @@ GetLinkedGenes <- function(
 #' CCAN that it belongs to in the second column.
 #' @param threshold Threshold for retaining a coaccessible site. Links with
 #' a value less than or equal to this threshold will be discarded.
+#' @param sep Separators to use for strings encoding genomic coordinates.
+#' First element is used to separate the chromosome from the coordinates, second
+#' element is used to separate the start from end coordinate.
 #'
 #' @export
 #' @importFrom GenomicRanges start end makeGRangesFromDataFrame
@@ -94,11 +97,16 @@ GetLinkedGenes <- function(
 #'
 #' @concept links
 #' @return Returns a \code{\link[GenomicRanges]{GRanges}} object
-ConnectionsToLinks <- function(conns, ccans = NULL, threshold = 0) {
+ConnectionsToLinks <- function(
+  conns,
+  ccans = NULL,
+  threshold = 0,
+  sep = c("-", "-")
+) {
   # add chromosome information
-  chr1 <- stri_split_fixed(str = conns$Peak1, pattern = "-")
+  chr1 <- stri_split_fixed(str = conns$Peak1, pattern = sep[[1]])
   conns$chr1 <- unlist(x = chr1)[3 * (seq_along(along.with = chr1)) - 2]
-  chr2 <- stri_split_fixed(str = conns$Peak2, pattern = "-")
+  chr2 <- stri_split_fixed(str = conns$Peak2, pattern = sep[[1]])
   conns$chr2 <- unlist(x = chr2)[3 * (seq_along(along.with = chr2)) - 2]
 
   # filter out trans-chr links
@@ -124,8 +132,8 @@ ConnectionsToLinks <- function(conns, ccans = NULL, threshold = 0) {
   }
 
   # extract genomic regions
-  coords.1 <- StringToGRanges(regions = conns$Peak1, sep = c("-", "-"))
-  coords.2 <- StringToGRanges(regions = conns$Peak2, sep = c("-", "-"))
+  coords.1 <- StringToGRanges(regions = conns$Peak1, sep = sep)
+  coords.2 <- StringToGRanges(regions = conns$Peak2, sep = sep)
   chr <- seqnames(x = coords.1)
 
   # find midpoints
@@ -187,6 +195,8 @@ ConnectionsToLinks <- function(conns, ccans = NULL, threshold = 0) {
 #' p-value equal or greater than this value will be removed from the output.
 #' @param score_cutoff Minimum absolute value correlation coefficient for a link
 #' to be retained
+#' @param gene.id Set to TRUE if genes in the expression assay are named
+#' using gene IDs rather than gene names.
 #' @param verbose Display messages
 #'
 #' @importFrom Seurat GetAssayData
@@ -227,6 +237,7 @@ LinkPeaks <- function(
   n_sample = 200,
   pvalue_cutoff = 0.05,
   score_cutoff = 0.05,
+  gene.id = FALSE,
   verbose = TRUE
 ) {
   if (!inherits(x = object[[peak.assay]], what = "ChromatinAssay")) {
@@ -274,14 +285,12 @@ LinkPeaks <- function(
   peaks.keep <- peakcounts > min.cells
   genes.keep <- genecounts > min.cells
   peak.data <- peak.data[peaks.keep, ]
-  if (is.null(x = genes.use)) {
-    expression.data <- expression.data[genes.keep, ]
-  } else {
+  if (!is.null(x = genes.use)) {
     genes.keep <- intersect(
       x = names(x = genes.keep[genes.keep]), y = genes.use
     )
-    expression.data <- expression.data[genes.keep, , drop = FALSE]
   }
+  expression.data <- expression.data[genes.keep, , drop = FALSE]
   if (verbose) {
     message(
       "Testing ",
@@ -292,7 +301,20 @@ LinkPeaks <- function(
     )
   }
   genes <- rownames(x = expression.data)
-  gene.coords.use <- gene.coords[gene.coords$gene_name %in% genes,]
+  if (gene.id) {
+    gene.coords.use <- gene.coords[gene.coords$gene_id %in% genes,]
+    gene.coords.use$gene_name <- gene.coords.use$gene_id
+  } else {
+    gene.coords.use <- gene.coords[gene.coords$gene_name %in% genes,]
+  }
+  if (length(x = gene.coords.use) == 0) {
+    stop("Could not find gene coordinates for requested genes")
+  }
+  if (length(x = gene.coords.use) < nrow(x = expression.data)) {
+    message("Found gene coordinates for ",
+            length(x = gene.coords.use),
+            " genes")
+  }
   peaks <- granges(x = object[[peak.assay]])
   peaks <- peaks[peaks.keep]
   peak_distance_matrix <- DistanceToTSS(
@@ -518,7 +540,12 @@ LinksToGRanges <- function(linkmat, gene.coords, sep = c("-", "-")) {
 #' @importFrom GenomicRanges resize
 #
 # @return Returns a sparse matrix
-DistanceToTSS <- function(peaks, genes, distance = 200000, sep = c("-", "-")) {
+DistanceToTSS <- function(
+  peaks,
+  genes,
+  distance = 200000,
+  sep = c("-", "-")
+  ) {
   tss <- resize(x = genes, width = 1, fix = 'start')
   genes.extended <- suppressWarnings(
     expr = Extend(

diff --git a/R/objects.R b/R/objects.R
@@ -1157,7 +1157,7 @@ merge.ChromatinAssay <- function(
       if (all(count_nonzero > 0)) {
         merged.counts <- RowMergeSparseMatrices(
           mat1 = all.counts[[1]],
-          mat2 = all.counts[[2:length(x = all.counts)]]
+          mat2 = all.counts[2:length(x = all.counts)]
         )
         reduced.ranges <- StringToGRanges(regions = rownames(x = merged.counts))
       } else {
@@ -1167,7 +1167,7 @@ merge.ChromatinAssay <- function(
       if (all(data_nonzero > 0)) {
         merged.data <- RowMergeSparseMatrices(
           mat1 = all.data[[1]],
-          mat2 = all.data[[2:length(x = all.data)]]
+          mat2 = all.data[2:length(x = all.data)]
         )
         reduced.ranges <- SetIfNull(
           x = reduced.ranges,

diff --git a/R/preprocessing.R b/R/preprocessing.R
@@ -552,6 +552,8 @@ RegionStats.Seurat <- function(
 #'  \eqn{IDF}.
 #' }
 #' @param scale.factor Which scale factor to use. Default is 10000.
+#' @param idf A precomputed IDF vector to use. If NULL, compute based on the
+#' input data matrix.
 #' @param verbose Print progress
 #' @rdname RunTFIDF
 #' @importFrom Matrix colSums rowSums Diagonal tcrossprod
@@ -566,6 +568,7 @@ RunTFIDF.default <- function(
   assay = NULL,
   method = 1,
   scale.factor = 1e4,
+  idf = NULL,
   verbose = TRUE,
   ...
 ) {
@@ -587,18 +590,41 @@ RunTFIDF.default <- function(
   } else {
     tf <- tcrossprod(x = object, y = Diagonal(x = 1 / npeaks))
   }
-  rsums <- rowSums(x = object)
-  if (any(rsums == 0)) {
-    warning("Some features contain 0 total counts")
+  if (!is.null(x = idf)) {
+    precomputed_idf <- TRUE
+    if (!inherits(x = idf, what = "numeric")) {
+      stop("idf parameter must be a numeric vector")
+    }
+    if (length(x = idf) != nrow(x = object)) {
+      stop("Length of supplied IDF vector does not match",
+           " number of rows in input matrix")
+    }
+    if (any(idf == 0)) {
+      stop("Supplied IDF values cannot be zero")
+    }
+    if (verbose) {
+      message("Using precomputed IDF vector")
+    }
+  } else {
+    precomputed_idf <- FALSE
+    rsums <- rowSums(x = object)
+    if (any(rsums == 0)) {
+      warning("Some features contain 0 total counts")
+    }
+    idf <- ncol(x = object) / rsums
   }
-  idf <- ncol(x = object) / rsums
+
   if (method == 2) {
-    idf <- log(1 + idf)
+    if (!precomputed_idf) {
+      idf <- log(1 + idf)
+    }
   } else if (method == 3) {
     slot(object = tf, name = "x") <- log1p(
       x = slot(object = tf, name = "x") * scale.factor
     )
-    idf <- log(1 + idf)
+    if (!precomputed_idf) {
+      idf <- log(1 + idf)
+    }
   }
   norm.data <- Diagonal(n = length(x = idf), x = idf) %*% tf
   if (method == 1) {
@@ -626,6 +652,7 @@ RunTFIDF.Assay <- function(
   assay = NULL,
   method = 1,
   scale.factor = 1e4,
+  idf = NULL,
   verbose = TRUE,
   ...
 ) {
@@ -634,6 +661,7 @@ RunTFIDF.Assay <- function(
     method = method,
     assay = assay,
     scale.factor = scale.factor,
+    idf = idf,
     verbose = verbose,
     ...
   )
@@ -658,6 +686,7 @@ RunTFIDF.Seurat <- function(
   assay = NULL,
   method = 1,
   scale.factor = 1e4,
+  idf = NULL,
   verbose = TRUE,
   ...
 ) {
@@ -668,6 +697,7 @@ RunTFIDF.Seurat <- function(
     assay = assay,
     method = method,
     scale.factor = scale.factor,
+    idf = idf,
     verbose = verbose,
     ...
   )