Added docs for bioconductor submission -- related to #96

jr-leary7 · Nov 12, 2023 · afe646c · afe646c
1 parent 93a0ed6
commit afe646c
Show file tree

Hide file tree

Showing 56 changed files with 317 additions and 412 deletions.
diff --git a/.github/workflows/bioc-check.yaml b/.github/workflows/bioc-check.yaml
@@ -4,7 +4,8 @@
 on:
   push:
     branches: main
-  pull_request: main
+  pull_request:
+    branches: main
 
 name: bioc-check
 

diff --git a/.github/workflows/render-README.yaml b/.github/workflows/render-README.yaml
@@ -10,7 +10,7 @@ name: render-README
 
 jobs:
   render:
-    runs-on: ubuntu-latest
+    runs-on: macos-latest
     env:
       GITHUB_PAT: ${{ secrets.GITHUB_TOKEN }}
     steps:
@@ -22,9 +22,9 @@ jobs:
       - uses: r-lib/actions/setup-pandoc@v2
 
       - name: install CRAN packages
-        run: Rscript -e 'install.packages(c("rmarkdown","ggplot2", "dplyr", "purrr", "remotes", "devtools", "BiocManager", "Seurat"), dependencies = TRUE)'
+        run: Rscript -e 'install.packages(c("rmarkdown","ggplot2", "dplyr", "purrr", "remotes", "devtools", "BiocManager", "Seurat"), force = TRUE)'
       - name: install BioConductor packages
-        run: Rscript -e 'BiocManager::install(c("SingleCellExperiment", "scater", "scran", "scuttle", "bluster"))'
+        run: Rscript -e 'BiocManager::install(c("SingleCellExperiment", "scater", "scran", "scuttle", "bluster"), force = TRUE)'
       - name: install GitHub packages
         run: Rscript -e 'remotes::install_github("jr-leary7/scLANE")'
       - name: render README

diff --git a/DESCRIPTION b/DESCRIPTION
@@ -63,6 +63,7 @@ Suggests:
     BiocStyle, 
     slingshot, 
     gprofiler2,
+    BiocParallel, 
     BiocGenerics, 
     BiocNeighbors, 
     testthat (>= 3.0.0),

diff --git a/R/GetResultsDE.R b/R/GetResultsDE.R
@@ -16,12 +16,8 @@
 #' @seealso \code{\link{testDynamic}}
 #' @seealso \code{\link[stats]{p.adjust}}
 #' @examples
-#' \dontrun{
-#' getResultsDE(gene_stats)
-#' getResultsDE(gene_stats,
-#'              p.adj.method = "BH",
-#'              fdr.cutoff = 5e-3)
-#' }
+#' data(scLANE_models)
+#' scLANE_de_res <- getResultsDE(scLANE_models)
 
 getResultsDE <- function(test.dyn.res = NULL,
                          p.adj.method = "holm",

diff --git a/R/clusterGenes.R b/R/clusterGenes.R
@@ -23,19 +23,12 @@
 #' @seealso \code{\link{plotClusteredGenes}}
 #' @export
 #' @examples
-#' \dontrun{
-#' clusterGenes(gene_stats, pt = pt_df)
-#' clusterGenes(gene_stats,
-#'              pt = pt_df,
-#'              size.factor.offset = createCellOffset(sce_obj),
-#'              clust.algo = "kmeans",
-#'              use.pca = TRUE,
-#'              n.PC = 10,
-#'              lineages = "B")
-#' clusterGenes(gene_stats,
-#'              pt = pt_df,
-#'              lineages = c("A", "C"))
-#' }
+#' data(sim_pseudotime)
+#' data(scLANE_models)
+#' cell_offset <- createCellOffset(sim_counts)
+#' gene_clusters <- clusterGenes(scLANE_models,
+#'                               pt = sim_pseudotime,
+#'                               size.factor.offset = cell_offset)
 
 clusterGenes <- function(test.dyn.res = NULL,
                          pt = NULL,

diff --git a/R/createCellOffset.R b/R/createCellOffset.R
@@ -13,11 +13,8 @@
 #' @seealso \code{\link[scuttle]{computeLibraryFactors}}
 #' @export
 #' @examples
-#' \dontrun{
-#' createCellOffset(expr.mat = sce_obj)
-#' createCellOffset(expr.mat = counts(sce_obj))
-#' createCellOffset(expr.mat = seu_obj, scale.factor = 1e5)
-#' }
+#' data(sim_counts)
+#' cell_offset <- createCellOffset(sim_counts)
 
 createCellOffset <- function(expr.mat = NULL, scale.factor = 1e4) {
   # check inputs

diff --git a/R/createSlopeTestData.R b/R/createSlopeTestData.R
@@ -11,13 +11,6 @@
 #' @return A data.frame containing model data.
 #' @seealso \code{\link{marge2}}
 #' @seealso \code{\link{testSlope}}
-#' @examples
-#' \dontrun{
-#' createSlopeTestData(marge_mod, pt_df)
-#' createSlopeTestData(marge_mod,
-#'                     pt = pt_df,
-#'                     is.glmm = TRUE)
-#' }
 
 createSlopeTestData <- function(marge.model = NULL,
                                 pt = NULL,

diff --git a/R/embedGenes.R b/R/embedGenes.R
@@ -17,14 +17,16 @@
 #' @param k.param (Optional) The value of nearest-neighbors used in creating the SNN graph prior to clustering & in running UMAP. Defaults to 20.
 #' @param resolution.param (Optional) The value of the resolution parameter for the Leiden algorithm. If unspecified, silhouette scoring is used to select an optimal value. Defaults to NULL.
 #' @param random.seed (Optional) The random seed used to control stochasticity in the clustering algorithm. Defaults to 312.
+#' @param n.cores (Optional) Integer specifying the number of threads used by \code{\link[uwot]{umap}} and in \code{\link[bluster]{makeSNNGraph}}. Defaults to 2.
 #' @return A data.frame containing embedding coordinates, cluster IDs, and metadata for each gene.
 #' @export
 #' @examples
-#' \dontrun{
-#' embedGenes(smoothed_counts$Lineage_A,
-#'            pcs.return = 3,
-#'            cluster.genes = TRUE)
-#' }
+#' data(sim_pseudotime)
+#' data(scLANE_models)
+#' smoothed_dynamics <- smoothedCountsMatrix(scLANE_models,
+#'                                           pt = sim_pseudotime,
+#'                                           n.cores = 1L)
+#' gene_embed <- embedGenes(smoothed_dynamics$Lineage_A, n.cores = 1L)
 
 embedGenes <- function(smoothed.counts = NULL,
                        genes = NULL,
@@ -35,7 +37,8 @@ embedGenes <- function(smoothed.counts = NULL,
                        gene.meta.data = NULL,
                        k.param = 20,
                        resolution.param = NULL,
-                       random.seed = 312) {
+                       random.seed = 312,
+                       n.cores = 2L) {
   # check inputs
   if (is.null(smoothed.counts)) { stop("You forgot to provide a smoothed counts matrix to embedGenes().") }
   genes <- colnames(smoothed.counts)
@@ -52,28 +55,32 @@ embedGenes <- function(smoothed.counts = NULL,
                                        n_neighbors = k.param,
                                        init = "spectral",
                                        nn_method = "annoy",
-                                       seed = random.seed)
+                                       seed = random.seed,
+                                       n_threads = n.cores)
   } else {
     smoothed_counts_umap <- uwot::umap(smoothed.counts,
                                        n_components = 2,
                                        metric = "cosine",
                                        n_neighbors = k.param,
                                        init = "spectral",
                                        nn_method = "annoy",
-                                       seed = random.seed)
+                                       seed = random.seed,
+                                       n_threads = n.cores)
   }
   # clustering w/ silhouette score parameter tuning
   if (cluster.genes) {
     if (pca.init) {
       smoothed_counts_snn <- bluster::makeSNNGraph(smoothed_counts_pca$x,
                                                    k = k.param,
                                                    type = "jaccard",
-                                                   BNPARAM = BiocNeighbors::AnnoyParam(distance = "Cosine"))
+                                                   BNPARAM = BiocNeighbors::AnnoyParam(distance = "Cosine"),
+                                                   BPPARAM = BiocParallel::SnowParam(workers = n.cores))
     } else {
       smoothed_counts_snn <- bluster::makeSNNGraph(smoothed.counts,
                                                    k = k.param,
                                                    type = "jaccard",
-                                                   BNPARAM = BiocNeighbors::AnnoyParam(distance = "Cosine"))
+                                                   BNPARAM = BiocNeighbors::AnnoyParam(distance = "Cosine"),
+                                                   BPPARAM = BiocParallel::SnowParam(workers = n.cores))
     }
     if (is.null(resolution.param)) {
       if (pca.init) {

diff --git a/R/enrichDynamicGenes.R b/R/enrichDynamicGenes.R
@@ -12,12 +12,9 @@
 #' @seealso \code{\link[gprofiler2]{gost}}
 #' @export
 #' @examples
-#' \dontrun{
-#' enrichDynamicGenes(scLANE.de.res = de_stats)
-#' enrichDynamicGenes(scLANE.de.res = de_stats,
-#'                    lineage = "B",
-#'                    species = "mmusculus")
-#' }
+#' data(scLANE_models)
+#' scLANE_de_res <- getResultsDE(scLANE_models)
+#' enr_res <- enrichDynamicGenes(scLANE_de_res)
 
 enrichDynamicGenes <- function(scLANE.de.res = NULL,
                                lineage = NULL,

diff --git a/R/extractBreakpoints.R b/R/extractBreakpoints.R
@@ -8,9 +8,13 @@
 #' @return A data.frame of breakpoints & their directions.
 #' @export
 #' @examples
-#' \dontrun{
-#' extractBreakpoints(model = marge_mod)
-#' }
+#' data(sim_counts)
+#' data(sim_pseudotime)
+#' cell_offset <- createCellOffset(sim_counts)
+#' marge_model <- marge2(sim_pseudotime,
+#'                       Y = BiocGenerics::counts(sim_counts)[4, ],
+#'                       Y.offset = cell_offset)
+#' breakpoint_df <- extractBreakpoints(model = marge_model)
 
 extractBreakpoints <- function(model = NULL, directions = TRUE) {
   # check inputs

diff --git a/R/fitGLMM.R b/R/fitGLMM.R
@@ -22,12 +22,13 @@
 #' @seealso \code{\link{modelLRT}}
 #' @export
 #' @examples
-#' \dontrun{
-#' fitGLMM(X_pred = pt_df,
-#'         Y = raw_counts,
-#'         Y.offset = size_factor_vec,
-#'         id.vec = subject_vec)
-#' }
+#' data(sim_counts)
+#' data(sim_pseudotime)
+#' cell_offset <- createCellOffset(sim_counts)
+#' glmm_mod <- fitGLMM(X_pred = sim_pseudotime,
+#'                     Y = BiocGenerics::counts(sim_counts)[4, ],
+#'                     Y.offset = cell_offset,
+#'                     id.vec = sim_counts$subject)
 
 fitGLMM <- function(X_pred = NULL,
                     Y = NULL,

diff --git a/R/geneProgramScoring.R b/R/geneProgramScoring.R
@@ -15,18 +15,22 @@
 #' @return Either a \code{Seurat} or \code{SingleCellExperiment} object if \code{expr.mat} is in either form, or a data.frame containing per-cell program scores if \code{expr.mat} is a matrix.
 #' @export
 #' @examples
-#' \dontrun{
-#' geneProgramScoring(seu_obj,
-#'                    genes = gene_embed$gene,
-#'                    gene.clusters = gene_embed$leiden,
-#'                    program.labels = c("cell cycle", "organogenesis"))
-#' }
+#' data(sim_pseudotime)
+#' data(scLANE_models)
+#' smoothed_dynamics <- smoothedCountsMatrix(scLANE_models,
+#'                                           pt = sim_pseudotime,
+#'                                           n.cores = 1L)
+#' gene_embed <- embedGenes(smoothed_dynamics$Lineage_A, n.cores = 1L)
+#' sim_counts <- geneProgramScoring(sim_counts,
+#'                                  genes = gene_embed$gene,
+#'                                  gene.clusters = gene_embed$leiden,
+#'                                  n.cores = 1L)
 
 geneProgramScoring <- function(expr.mat = NULL,
                                genes = NULL,
                                gene.clusters = NULL,
                                program.labels = NULL,
-                               n.cores = 2) {
+                               n.cores = 2L) {
   # check inputs
   if (is.null(expr.mat) || is.null(genes) || is.null(gene.clusters)) { stop("Arguments to geneProgramScoring() are missing.") }
   if (!is.factor(gene.clusters)) {

diff --git a/R/getFittedValues.R b/R/getFittedValues.R
@@ -20,14 +20,13 @@
 #' @return A data.frame containing depth- and log1p-normalized expression, model predictions, and cell-level metadata.
 #' @export
 #' @examples
-#' \dontrun{
-#' getFittedValues(gene_stats,
-#'                 genes = c("Neurog3", "Epcam", "Krt19"),
-#'                 pt = pt_df,
-#'                 expr.mat = seu_obj,
-#'                 cell.meta.data = [email protected],
-#'                 ci.alpha = 0.05)
-#' }
+#' data(sim_counts)
+#' data(sim_pseudotime)
+#' data(scLANE_models)
+#' fitted_vals <- getFittedValues(scLANE_models,
+#'                                genes = sample(names(scLANE_models), 5),
+#'                                pt = sim_pseudotime,
+#'                                expr.mat = sim_counts)
 
 getFittedValues <- function(test.dyn.res = NULL,
                             genes = NULL,

diff --git a/R/getKnotDist.R b/R/getKnotDist.R
@@ -11,9 +11,8 @@
 #' @return A data.frame containing gene name, lineage ID, and knot location in pseudotime.
 #' @export
 #' @examples
-#' \dontrun{
-#' getKnotDist(gene_stats)
-#' }
+#' data(scLANE_models)
+#' knot_dist <- getKnotDist(scLANE_models)
 
 getKnotDist <- function(test.dyn.res = NULL, dyn.genes = NULL) {
   # check inputs

diff --git a/R/marge2.R b/R/marge2.R
@@ -42,24 +42,12 @@
 #' @seealso \code{\link[geeM]{geem}}
 #' @export
 #' @examples
-#' \dontrun{
-#'   marge2(pseudotime_df,
-#'          Y = expr_vec,
-#'          M = 3)
-#'   marge2(pseudotime_df,
-#'          Y = expr_vec,
-#'          Y.offset = size_factor_vec,
-#'          is.gee = TRUE,
-#'          id.vec = subject_vec,
-#'          cor.structure = "exchangeable")
-#'   marge2(pseudotime_df,
-#'          Y = expr_vec,
-#'          is.gee = TRUE,
-#'          id.vec = subject_vec,
-#'          cor.structure = "ar1",
-#'          n.knot.max = 10,
-#'          return.basis = TRUE)
-#' }
+#' data(sim_counts)
+#' data(sim_pseudotime)
+#' cell_offset <- createCellOffset(sim_counts)
+#' marge_model <- marge2(sim_pseudotime,
+#'                       Y = BiocGenerics::counts(sim_counts)[4, ],
+#'                       Y.offset = cell_offset)
 
 marge2 <- function(X_pred = NULL,
                    Y = NULL,

diff --git a/R/nbGAM.R b/R/nbGAM.R
@@ -21,18 +21,14 @@
 #' @seealso \code{\link[gamlss.dist]{NBI}}
 #' @export
 #' @examples
-#' \dontrun{
-#' nbGAM(expr_vec, pt_df)
-#' nbGAM(expr_vec,
-#'       pt = pt_df,
-#'       id.vec = subject_ids,
-#'       random.slopes = TRUE)
-#' nbGAM(expr_vec,
-#'       pt = pt_df,
-#'       Y.offset = size_factor_vec,
-#'       penalize.spline = TRUE,
-#'       spline.df = 10)
-#' }
+#' data(sim_counts)
+#' data(sim_pseudotime)
+#' cell_offset <- createCellOffset(sim_counts)
+#' gam_mod <- nbGAM(BiocGenerics::counts(sim_counts)[4, ],
+#'                  pt = sim_pseudotime,
+#'                  Y.offset = cell_offset,
+#'                  penalize.spline = TRUE,
+#'                  spline.df = 10)
 
 nbGAM <- function(expr = NULL,
                   pt = NULL,

diff --git a/R/npConvolve.R b/R/npConvolve.R
@@ -9,12 +9,12 @@
 #' @return A convolution with same length as the input vector.
 #' @details
 #' \itemize{
-#' \item The convolution here uses \code{\link[stats]{convolve}}, but creates the kernel and padding in such a way that it matches the output from `np.convolve` in Python's \code{numpy} matrix algebra package.
+#' \item The convolution here uses \code{\link[stats]{convolve}}, but creates the kernel and padding in such a way that it matches the output from \code{np.convolve} in Python's \code{numpy} matrix algebra package.
 #' }
 #' @seealso \code{\link[stats]{convolve}}
 #' @export
 #' @examples
-#' npConvolve(x = rnorm(20), conv.kernel = rep(1/5, 5))
+#' convolved_vec <- npConvolve(x = rnorm(20), conv.kernel = rep(1/5, 5))
 #'
 
 npConvolve <- function(x = NULL, conv.kernel = NULL) {

diff --git a/R/plotClusteredGenes.R b/R/plotClusteredGenes.R
@@ -37,7 +37,7 @@ plotClusteredGenes <- function(test.dyn.res = NULL,
                                pt = NULL,
                                size.factor.offset = NULL,
                                parallel.exec = TRUE,
-                               n.cores = 2) {
+                               n.cores = 2L) {
   # check inputs
   if (is.null(test.dyn.res) || is.null(gene.clusters) || is.null(pt)) { stop("Arguments to plotClusteredGenes() are missing.") }
   colnames(pt) <- paste0("Lineage_", LETTERS[seq_len(ncol(pt))])

diff --git a/R/plotModelCoefs.R b/R/plotModelCoefs.R
@@ -18,13 +18,16 @@
 #' @return A \code{ggplot2} object displaying a gene dynamics plot & a table of coefficients across pseudotime intervals.
 #' @export
 #' @examples
-#' \dontrun{
-#' plotModelCoefs(gene_stats,
-#'                gene = "BRCA2",
-#'                pt = pt_df,
-#'                expr.mat = seu_obj,
+#' data(sim_counts)
+#' data(sim_pseudotime)
+#' data(scLANE_models)
+#' cell_offset <- createCellOffset(sim_counts)
+#' scLANE_de_res <- getResultsDE(scLANE_models)
+#' plotModelCoefs(scLANE_models,
+#'                gene = scLANE_de_res$Gene[1],
+#'                pt = sim_pseudotime,
+#'                expr.mat = sim_counts,
 #'                size.factor.offset = cell_offset)
-#' }
 
 plotModelCoefs <- function(test.dyn.res = NULL,
                            gene = NULL,