Merge branch 'develop'

Author: timoast

.github/workflows/cmd-check.yml

@@ -38,7 +38,7 @@ env:
   run_covr: 'false'
   run_pkgdown: 'false'
   has_RUnit: 'false'
-  cache-version: 'cache-v1'
+  cache-version: 'cache-v3'
   run_docker: 'false'
 
 jobs:
@@ -54,7 +54,7 @@ jobs:
       fail-fast: false
       matrix:
         config:
-          - { os: ubuntu-latest, r: '4.0', bioc: '3.12', cont: "bioconductor/bioconductor_docker:RELEASE_3_12", rspm: "https://packagemanager.rstudio.com/cran/__linux__/focal/latest" }
+          - { os: ubuntu-latest, r: '4.1', bioc: '3.14', cont: "bioconductor/bioconductor_docker:RELEASE_3_14", rspm: "https://packagemanager.rstudio.com/cran/__linux__/focal/latest" }
 
     env:
       R_REMOTES_NO_ERRORS_FROM_WARNINGS: true

DESCRIPTION

@@ -1,7 +1,7 @@
 Package: Signac
 Title: Analysis of Single-Cell Chromatin Data
-Version: 1.5.0
-Date: 2021-12-07
+Version: 1.6.0
+Date: 2022-03-04
 Authors@R: c(
   person(given = 'Tim', family = 'Stuart', email = '[email protected]', role = c('aut', 'cre'), comment = c(ORCID = '0000-0002-3044-0897')),
   person(given = 'Avi', family = 'Srivastava', email = '[email protected]', role = 'aut', comment = c(ORCID = '0000-0001-9798-2079')),
@@ -30,7 +30,7 @@ Imports:
     Matrix,
     Rsamtools,
     S4Vectors,
-    Seurat (>= 4.0.0),
+    Seurat (>= 4.0.6),
     SeuratObject (>= 4.0.0),
     data.table,
     dplyr (>= 1.0.0),
@@ -54,7 +54,8 @@ Imports:
     Rcpp,
     ggforce,
     qlcMatrix,
-    grid
+    grid,
+    tidyselect
 Collate: 
     'RcppExports.R'
     'data.R'
@@ -65,6 +66,7 @@ Collate:
     'fragments.R'
     'genomeinfodb-methods.R'
     'granges-methods.R'
+    'heatmaps.R'
     'iranges-methods.R'
     'links.R'
     'mito.R'
@@ -73,6 +75,7 @@ Collate:
     'peaks.R'
     'preprocessing.R'
     'quantification.R'
+    'region-enrichment.R'
     'utilities.R'
     'visualization.R'
     'zzz.R'

NAMESPACE

@@ -54,6 +54,9 @@ S3method(Links,ChromatinAssay)
 S3method(Links,Seurat)
 S3method(Motifs,ChromatinAssay)
 S3method(Motifs,Seurat)
+S3method(RegionMatrix,ChromatinAssay)
+S3method(RegionMatrix,Seurat)
+S3method(RegionMatrix,default)
 S3method(RegionStats,ChromatinAssay)
 S3method(RegionStats,Seurat)
 S3method(RegionStats,default)
@@ -145,12 +148,16 @@ export(LinkPlot)
 export(Links)
 export(LookupGeneCoords)
 export(MatchRegionStats)
+export(MotifCounts)
 export(MotifPlot)
 export(Motifs)
 export(NucleosomeSignal)
 export(PeakPlot)
 export(PlotFootprint)
 export(ReadMGATK)
+export(RegionHeatmap)
+export(RegionMatrix)
+export(RegionPlot)
 export(RegionStats)
 export(RunChromVAR)
 export(RunSVD)
@@ -303,6 +310,7 @@ importFrom(Seurat,Project)
 importFrom(Seurat,RenameCells)
 importFrom(Seurat,RowMergeSparseMatrices)
 importFrom(Seurat,SetAssayData)
+importFrom(Seurat,SetQuantile)
 importFrom(Seurat,VariableFeatures)
 importFrom(Seurat,WhichCells)
 importFrom(SeuratObject,Cells)
@@ -340,9 +348,11 @@ importFrom(ggplot2,guide_legend)
 importFrom(ggplot2,guides)
 importFrom(ggplot2,labs)
 importFrom(ggplot2,scale_color_gradient2)
+importFrom(ggplot2,scale_color_grey)
 importFrom(ggplot2,scale_color_identity)
 importFrom(ggplot2,scale_color_manual)
 importFrom(ggplot2,scale_fill_gradient)
+importFrom(ggplot2,scale_fill_grey)
 importFrom(ggplot2,scale_fill_manual)
 importFrom(ggplot2,scale_fill_viridis_c)
 importFrom(ggplot2,scale_x_continuous)
@@ -393,6 +403,7 @@ importFrom(stats,sd)
 importFrom(stringi,stri_split_fixed)
 importFrom(tidyr,pivot_longer)
 importFrom(tidyr,separate)
+importFrom(tidyselect,all_of)
 importFrom(tools,file_ext)
 importFrom(tools,md5sum)
 importFrom(utils,globalVariables)

NEWS.md

@@ -1,3 +1,27 @@
+# Signac 1.6.0
+
+New functionality:
+
+* Added `RegionHeatmap()` function
+* Added `RegionPlot()` function
+* Added `RegionMatrix()` function
+* Added `MotifCounts()` function
+* Added ability to plot multiple assays simultaneously in `CoveragePlot()` by providing a list of assay names
+* Added `assay.scale`, `bigwig.scale`, and `split.assay` parameters to `CoveragePlot()`
+* Added new option to provide quantile `ymax` value in `CoveragePlot()`
+* Added `bigwig.scale` and `ymax` parameters to `BigwigTrack()`
+* Added `peak.slot` parameter to `LinkPeaks()` ([#932](https://github.com/timoast/signac/issues/932))
+
+Bug fixes:
+
+* Fixed behaviour of `min.features` parameter in `CreateChromatinAssay()` to retain cells with `>= min.features` (#[902](https://github.com/timoast/signac/discussions/902))
+* Fixed behaviour of `min.cells` parameter in `LinkPeaks()` ([#932](https://github.com/timoast/signac/issues/932))
+
+Other changes:
+
+* Updated documentation for `CoveragePlot()` ([#992](https://github.com/timoast/signac/issues/992))
+* Removed `method` parameter from `LinkPeaks()` ([#932](https://github.com/timoast/signac/issues/932))
+
 # Signac 1.5.0
 
 Bug fixes:
@@ -17,6 +41,10 @@ Other changes:
 
 # Signac 1.4.0
 
+New functionality:
+
+* Added ability to display multiple assays in `CoveragePlot()`. The `assay` parameter can now be a list of assays to plot data from, with signal colored by assay of origin.
+
 Bug fixes:
 
 * Fixed bug in `FindMotifs()` when using only one region as input ([#732](https://github.com/timoast/signac/issues/732))

R/footprinting.R

@@ -495,27 +495,27 @@ RunFootprint <- function(
   if (verbose) {
     message("Computing observed Tn5 insertions per base")
   }
+  dna.sequence <- Biostrings::getSeq(x = genome, Extend(
+    x = regions,
+    upstream = 3,
+    downstream = 3
+    )
+  )
   if (compute.expected) {
     bias <- GetAssayData(object = object, slot = "bias")
     if (is.null(x = bias)) {
       stop("Insertion bias not computed")
     } else {
       # add three bases each side here so we can get the hexamer frequencies
       # for every position
-      dna.sequence <- Biostrings::getSeq(x = genome, Extend(
-        x = regions,
-        upstream = 3,
-        downstream = 3
-       )
-      )
       expected.insertions <- FindExpectedInsertions(
         dna.sequence = dna.sequence,
         bias = bias,
         verbose = verbose
       )
     }
   } else {
-    expected.insertions <- rep(1, width(x = dna.sequence)[[1]])
+    expected.insertions <- rep(1, width(x = dna.sequence)[[1]] - 6)
   }
 
   # count insertions at each position for each cell

R/fragments.R

@@ -32,7 +32,9 @@ head.Fragment <- function(x, n = 6L, ...) {
 #'
 #' Count total fragments per cell barcode present in a fragment file.
 #'
-#' @param fragments Path to a fragment file
+#' @param fragments Path to a fragment file. If a list of fragment files is
+#' provided, the total fragments for each cell barcode across all files will be
+#' returned
 #' @param cells Cells to include. If NULL, include all cells
 #' @param max_lines Maximum number of lines to read from the fragment file. If
 #' NULL, read all lines in the file.
@@ -41,7 +43,14 @@ head.Fragment <- function(x, n = 6L, ...) {
 #' @rdname CountFragments
 #' @export
 #' @concept fragments
-#' @return Returns a data.frame
+#' @return Returns a data.frame with the following columns:
+#' \itemize{
+#'   \item{CB: the cell barcode}
+#'   \item{frequency_count: total number of fragments sequenced for the cell}
+#'   \item{mononucleosome: total number of fragments with length between 147 bp and 294 bp}
+#'   \item{nucleosome_free: total number of fragments with length <147 bp}
+#'   \item{reads_count: total number of reads sequenced for the cell}
+#' }
 #' @examples
 #' fpath <- system.file("extdata", "fragments.tsv.gz", package="Signac")
 #' counts <- CountFragments(fragments = fpath)
@@ -51,22 +60,45 @@ CountFragments <- function(
   max_lines = NULL,
   verbose = TRUE
 ) {
-  if (isRemote(x = fragments)) {
-    stop("Remote fragment files not supported")
+  if (!inherits(x = fragments, what = "list")) {
+    fragments <- list(fragments)
   }
-  fragments <- normalizePath(path = fragments, mustWork = TRUE)
-  max_lines <- SetIfNull(x = max_lines, y = 0)
-  verbose = as.logical(x = verbose)
-  if (!is.null(x = cells)) {
-    cells <- unique(x = cells)
+  for (i in seq_along(along.with = fragments)) {
+    if (isRemote(x = i)) {
+      stop("Remote fragment files not supported")
+    }
+    fragments[[i]] <- normalizePath(path = fragments[[i]], mustWork = TRUE)
+    max_lines <- SetIfNull(x = max_lines, y = 0)
+    verbose = as.logical(x = verbose)
+    if (!is.null(x = cells)) {
+      cells <- unique(x = cells)
+    }
+    counts <- groupCommand(
+      fragments = fragments[[i]],
+      some_whitelist_cells = cells,
+      max_lines = max_lines,
+      verbose = verbose
+    )
+    rownames(x = counts) <- counts$CB
+    counts$CB <- NULL
+    if (i == 1) {
+      # first file
+      allcounts <- counts
+    } else {
+      # merge
+      common <- intersect(
+        x = rownames(x = allcounts), y = rownames(x = counts)
+      )
+      allcounts[common, ] <- allcounts[common, ] + counts[common, ]
+      missing_cells <- setdiff(x = rownames(x = counts), y = common)
+      allcounts <- rbind(allcounts, counts[missing_cells, ])
+    }
   }
-  counts <- groupCommand(
-    fragments = fragments,
-    some_whitelist_cells = cells,
-    max_lines = max_lines,
-    verbose = verbose
-  )
-  return(counts)
+  # reformat for backwards compatibility
+  allcounts$CB <- rownames(x = allcounts)
+  rownames(x = allcounts) <- NULL
+  allcounts <- allcounts[, c(5, 1, 2, 3, 4)]
+  return(allcounts)
 }
 
 #' Filter cells from fragment file

R/generics.R

@@ -270,6 +270,20 @@ IdentifyVariants <- function(object, ...) {
   UseMethod(generic = "IdentifyVariants", object = object)
 }
 
+#' Region enrichment analysis
+#'
+#' Count fragments within a set of regions for different groups of
+#' cells.
+#'
+#' @param object A Seurat or ChromatinAssay object
+#' @param ... Arguments passed to other methods
+#' @return Returns a \code{\link[SeuratObject]{Seurat}} object
+#' @rdname RegionMatrix
+#' @export RegionMatrix
+RegionMatrix <- function(object, ...) {
+  UseMethod(generic = "RegionMatrix", object = object)
+}
+
 #' Compute base composition information for genomic ranges
 #'
 #' Compute the GC content, region lengths, and dinucleotide base frequencies