Merge pull request #694 from joshua-d-campbell/master

Merging from devel for new release 2.8.1

Author: joshua-d-campbell

.Rbuildignore

@@ -15,3 +15,5 @@ exec/png
 .covrignore
 ^pkgdown$
 ^\.github$
+^vignettes/articles/*
+^images

.github/workflows/BioC-check.yaml

@@ -39,9 +39,16 @@ jobs:
         if: runner.os == 'macOS'
         run: brew install fftw
 
-      - uses: r-lib/actions/setup-r-dependencies@v1
+      - name: Manual Install
+        run: |
+          install.packages("BiocManager")
+          BiocManager::install("ggtree")
+        shell: Rscript {0}
+
+      - uses: r-lib/actions/setup-r-dependencies@v2
         with:
           extra-packages: rcmdcheck
+          cache-version: 2
 
       - name: Install XQuartz on macOS
         if: runner.os == 'macOS'

.github/workflows/R-CMD-check.yaml

@@ -27,12 +27,12 @@ jobs:
       R_KEEP_PKG_SOURCE: yes
 
     steps:
-      - uses: actions/checkout@v2
+      - uses: actions/checkout@v3
 
-      - uses: r-lib/actions/setup-pandoc@v1
+      - uses: r-lib/actions/setup-pandoc@v2
 
       - name: Setup Python
-        uses: actions/setup-python@v2.2.2
+        uses: actions/setup-python@v2
         with:
           python-version: '3.8'
 
@@ -57,7 +57,7 @@ jobs:
           which python
         shell: bash
 
-      - uses: r-lib/actions/setup-r@v1
+      - uses: r-lib/actions/setup-r@v2
         with:
           r-version: ${{ matrix.config.r }}
           http-user-agent: ${{ matrix.config.http-user-agent }}
@@ -71,16 +71,18 @@ jobs:
         if: runner.os == 'macOS'
         run: brew install fftw
 
-      - uses: r-lib/actions/setup-r-dependencies@v1
-        with:
-          extra-packages: rcmdcheck
-
-      - name: Install Celda devel
+      - name: Manual Install
         run: |
-          install.packages("devtools")
-          devtools::install_github("campbio/celda@devel")
+          install.packages("BiocManager")
+          BiocManager::install("ggtree")
         shell: Rscript {0}
 
+      - uses: r-lib/actions/setup-r-dependencies@v2
+        with:
+          cache-version: 2
+          extra-packages: any::rcmdcheck
+          needs: check
+
       - name: Install XQuartz on macOS
         if: runner.os == 'macOS'
         run: brew install xquartz --cask

DESCRIPTION

@@ -1,14 +1,15 @@
 Package: singleCellTK
 Type: Package
 Title: Comprehensive and Interactive Analysis of Single Cell RNA-Seq Data
-Version: 2.8.0
+Version: 2.8.1
 Authors@R: c(person(given="Yichen", family="Wang", email="wangych@bu.edu", role=c("aut", "cre"),
                     comment = c(ORCID = "0000-0003-4347-5199")),
              person(given="Irzam", family="Sarfraz", email="isarfraz@bu.edu", role=c("aut"),
                     comment = c(ORCID = "0000-0001-8121-792X")),
              person(given="Rui", family="Hong", email="rzhong@bu.edu", role=c("aut")),
              person(given="Yusuke", family="Koga", email="ykoga07@bu.edu", role=c("aut")),
              person(given="Salam", family="Alabdullatif", email="salamha@bu.edu", role=c("aut")),
+             person(given="Nida", family="Pervaiz", email="nidapervaiz27@gmail.com", role=c("aut")),
              person(given="David", family="Jenkins", email="dfj@bu.edu", role=c("aut"),
                     comment = c(ORCID = "0000-0002-7451-4288")),
              person(given="Vidya", family="Akavoor", email="vidyaap@bu.edu", role=c("aut")),
@@ -64,7 +65,7 @@ Imports:
     KernSmooth,
     limma,
     MAST,
-    Matrix,
+    Matrix (>= 1.5-3),
     matrixStats,
     methods,
     msigdbr,
@@ -94,7 +95,7 @@ Imports:
     reticulate (>= 1.14),
     tools,
     tximport,
-    fishpond,
+    eds,
     withr,
     GSEABase,
     R.utils,
@@ -113,8 +114,9 @@ Imports:
     TrajectoryUtils,
     scuttle,
     utils,
-    stats
-RoxygenNote: 7.2.1
+    stats,
+    zellkonverter
+RoxygenNote: 7.2.3
 Suggests:
     testthat,
     Rsubread,

NAMESPACE

@@ -111,6 +111,20 @@ export(plotSCEViolin)
 export(plotSCEViolinAssayData)
 export(plotSCEViolinColData)
 export(plotScDblFinderResults)
+export(plotScanpyDotPlot)
+export(plotScanpyEmbedding)
+export(plotScanpyHVG)
+export(plotScanpyHeatmap)
+export(plotScanpyMarkerGenes)
+export(plotScanpyMarkerGenesDotPlot)
+export(plotScanpyMarkerGenesHeatmap)
+export(plotScanpyMarkerGenesMatrixPlot)
+export(plotScanpyMarkerGenesViolin)
+export(plotScanpyMatrixPlot)
+export(plotScanpyPCA)
+export(plotScanpyPCAGeneRanking)
+export(plotScanpyPCAVariance)
+export(plotScanpyViolin)
 export(plotScdsHybridResults)
 export(plotScrubletResults)
 export(plotSeuratElbow)
@@ -182,6 +196,14 @@ export(runQuickUMAP)
 export(runSCANORAMA)
 export(runSCMerge)
 export(runScDblFinder)
+export(runScanpyFindClusters)
+export(runScanpyFindHVG)
+export(runScanpyFindMarkers)
+export(runScanpyNormalizeData)
+export(runScanpyPCA)
+export(runScanpyScaleData)
+export(runScanpyTSNE)
+export(runScanpyUMAP)
 export(runScranSNN)
 export(runScrublet)
 export(runSeuratFindClusters)
@@ -245,7 +267,7 @@ import(DelayedArray)
 import(DropletUtils)
 import(GSVAdata)
 import(SingleCellExperiment)
-import(fishpond)
+import(eds)
 importFrom(BiocParallel,SerialParam)
 importFrom(S4Vectors,"metadata<-")
 importFrom(S4Vectors,metadata)

NEWS.md

@@ -1,3 +1,13 @@
+Changes in Version 2.8.1 (2022-03-10)
+================================================================================
+* Added scanpy wrapper functions for use from console
+* Added scanpy UI curated workflow
+* Integrated scanpy to a la carte workflow
+* Fixed a bug in importing fluidigm dataset
+* Updated downloading features in the Shiny app
+* Added error checking around Enrichr functions
+* Minor tweaks to plot defaults
+
 Changes in Version 2.8.0 (2022-12-19)
 ================================================================================
 * Updated version to match Bioconductor version

R/enrichRSCE.R

@@ -85,7 +85,7 @@ runEnrichR <- function(inSCE,
   internetConnection <- suppressWarnings(Biobase::testBioCConnection())
   #check for internet connection
   if (!internetConnection){
-    stop("Please connect to the Internet and continue..")
+    stop("The Enrichr database could not be queried. No internet connection available.")
   }
   err <- tryCatch(
     {
@@ -95,38 +95,42 @@ runEnrichR <- function(inSCE,
   )
   #options(enrichR.base.address = "https://maayanlab.cloud/Enrichr/")
   #options(enrichR.live = TRUE)
-  temp_db <- enrichR::listEnrichrDbs()
+  temp_db <- NULL
+  err <- tryCatch(
+    {
+      temp_db <- .getEnrichrDB(noConnect = "error")
+    },
+    error = function(e) {}
+  )
   enrdb <- temp_db$libraryName
   #test for db existing
   if (is.null(db)){
     db <- enrdb
   } else if (!all(db %in% enrdb)){
     db.notFound <- db[!db %in% enrdb]
-    stop("database ", paste(db.notFound, collapse = ", "), " do not exist.")
+    stop("Database(s) ", paste(db.notFound, collapse = ", "), " were not found in Enrichr.")
   }
 
   enriched <- enrichR::enrichr(features, db)
   enriched <- data.frame(data.table::rbindlist(enriched, use.names = TRUE,
                                                fill = TRUE,
                                                idcol = "Database_selected"))
 
-  enriched$link <- vapply(enriched$Database_selected, function(x){
-    temp_db$link[temp_db$libraryName %in% x]
-  }, FUN.VALUE = character(1))
-  
-  #sort the results based on p-values
-  enriched <- enriched[order(enriched$P.value, decreasing = FALSE), ]
-  
-  #round the numeric values to their 7th digit
-  #nums <- vapply(enriched, is.numeric, FUN.VALUE = logical(1))
-  #enriched[, nums] <- round(enriched[, nums], digits = 7)
+  if(!is.null(enriched) && nrow(enriched) > 0) {
+
+    #sort the results based on p-values
+    enriched <- enriched[order(enriched$P.value, decreasing = FALSE), ]
+    
+    getEnrichRResult(inSCE, analysisName) <- list(result = enriched,
+                                                  param = list(
+                                                    features = features,
+                                                    by = by,
+                                                    db = db
+                                                  )) 
+  } else {
+    stop("No enrichments identified. No enrichment results were created.")
+  }
 
-  getEnrichRResult(inSCE, analysisName) <- list(result = enriched,
-                                                param = list(
-                                                  features = features,
-                                                  by = by,
-                                                  db = db
-                                                )) 
   return(inSCE)
 }
 
@@ -181,3 +185,17 @@ setReplaceMethod("getEnrichRResult",
                    S4Vectors::metadata(inSCE)$sctk$runEnrichR[[analysisName]] <- value
                    return(inSCE)
                  })
+
+.getEnrichrDB <- function(noConnect = c("warn", "error", "ignore")) {
+  db <- NULL
+  noConnect <- match.arg(noConnect)
+  db <- enrichR::listEnrichrDbs()
+  if(is.null(db)) {
+    if(noConnect == "error") {
+      stop("Connection to the online Enrichr database could not be established. No database information retrieved.")
+    } else if (noConnect == "warn") {
+      warning("Connection to the online Enrichr database could not be established. No database information retrieved.")
+    }
+  }
+  return(db)
+}

R/getTopHVG.R

@@ -11,7 +11,8 @@
 #' @param inSCE Input \linkS4class{SingleCellExperiment} object
 #' @param method Specify which method to use for variable gene extraction
 #' from Seurat \code{"vst"}, \code{"mean.var.plot"}, \code{"dispersion"} or
-#' Scran \code{"modelGeneVar"}. Default \code{"vst"}
+#' Scran \code{"modelGeneVar"} or Scanpy \code{"seurat"}, \code{"cell_ranger"}, 
+#' \code{"seurat_v3"}. Default \code{"vst"}
 #' @param hvgNumber Specify the number of top variable genes to extract.
 #' @param useFeatureSubset Get the feature names in the HVG list set by
 #' \code{setTopHVG}. Will ignore \code{method} and \code{hvgNumber} if not
@@ -52,7 +53,8 @@
 #' @importFrom S4Vectors metadata
 getTopHVG <- function(inSCE,
                       method = c("vst", "dispersion",
-                                 "mean.var.plot", "modelGeneVar"),
+                                 "mean.var.plot", "modelGeneVar", "seurat", 
+                                 "seurat_v3", "cell_ranger"),
                       hvgNumber = 2000,
                       useFeatureSubset = NULL,
                       featureDisplay = metadata(inSCE)$featureDisplay) {
@@ -64,7 +66,12 @@ getTopHVG <- function(inSCE,
     } else {
         metrics <- .dfFromHVGMetric(inSCE, method)
         metrics <- metrics[order(-metrics$v_rank),]
-        metrics <- metrics[metrics$v_rank > 0, ]
+        if (method == "seurat"){
+          metrics <- metrics[metrics$v_plot > 0, ]
+        }
+        else{
+          metrics <- metrics[metrics$v_rank > 0, ]
+        }
         if (method == "mean.var.plot") {
             means.use <- (metrics$mean > 0.1) & (metrics$mean < 8)
             dispersions.use <- (metrics$v_plot > 1) & (metrics$v_plot < Inf)
@@ -87,7 +94,8 @@ getTopHVG <- function(inSCE,
 #' @importFrom S4Vectors metadata<-
 setTopHVG <- function(inSCE,
                       method =  c("vst", "dispersion",
-                                  "mean.var.plot", "modelGeneVar"),
+                                  "mean.var.plot", "modelGeneVar", "seurat", 
+                                  "seurat_v3", "cell_ranger"),
                       hvgNumber = 2000,
                       featureSubsetName = NULL,
                       genes = NULL, genesBy = NULL,
@@ -138,13 +146,15 @@ setTopHVG <- function(inSCE,
 #' @param inSCE Input \linkS4class{SingleCellExperiment} object
 #' @param method Specify which method to use for variable gene extraction
 #' from Seurat \code{"vst"}, \code{"mean.var.plot"}, \code{"dispersion"} or
-#' Scran \code{"modelGeneVar"}. Default \code{"vst"}
+#' Scran \code{"modelGeneVar"} or Scanpy \code{"seurat"}, \code{"cell_ranger"}, 
+#' \code{"seurat_v3"}. Default \code{"vst"}
 #' @return data.frame object of HVG metrics calculated by \code{method},
 #' containing columns of \code{"mean"}, \code{"v_rank"}, \code{"v_plot"}
 #' @noRd
 .dfFromHVGMetric <- function(inSCE,
                              method = c("vst", "mean.var.plot", "dispersion",
-                                        "modelGeneVar")) {
+                                        "modelGeneVar", "seurat", 
+                                        "seurat_v3", "cell_ranger")) {
     method <- match.arg(method)
     df <- data.frame(featureNames = rownames(inSCE))
     if (method == "vst") {
@@ -167,6 +177,39 @@ setTopHVG <- function(inSCE,
                  "Run `runSeuratFindHVG()` with 'dispersion' method ",
                  "before using this function!")
         }
+    }else if (method == "seurat_v3") {
+      m <- "means"
+      v_rank <- "variances"
+      v_plot <- "variances_norm"
+      if (is.null(rowData(inSCE)[[v_rank]]) ||
+          is.null(rowData(inSCE)[[m]]) ||
+          is.null(rowData(inSCE)[[v_plot]])) {
+        stop("Scanpy variance metric not found in inSCE. ",
+             "Run `runScanpyFindHVG()` with 'seurat_v3' method ",
+             "before using this function!")
+      }
+    }else if (method == "cell_ranger") {
+      m <- "means"
+      v_rank <- "dispersions"
+      v_plot <- "dispersions_norm"
+      if (is.null(rowData(inSCE)[[v_rank]]) ||
+          is.null(rowData(inSCE)[[m]]) ||
+          is.null(rowData(inSCE)[[v_plot]])) {
+        stop("Scanpy dispersion metric not found in inSCE. ",
+             "Run `runScanpyFindHVG()` with 'cell_ranger' method ",
+             "before using this function!")
+      }
+    }else if (method == "seurat") {
+      m <- "means"
+      v_rank <- "dispersions"
+      v_plot <- "dispersions_norm"
+      if (is.null(rowData(inSCE)[[v_rank]]) ||
+          is.null(rowData(inSCE)[[m]]) ||
+          is.null(rowData(inSCE)[[v_plot]])) {
+        stop("Scanpy dispersion metric not found in inSCE. ",
+             "Run `runScanpyFindHVG()` with 'dispersion' method ",
+             "before using this function!")
+      }
     } else if (method == "modelGeneVar") {
         m <- "scran_modelGeneVar_mean"
         v_rank <- "scran_modelGeneVar_bio"