In bioinfocz/scdrake: A pipeline for droplet-based single-cell RNA-seq data secondary analysis implemented in the drake Make-like toolkit for R language

suppressPackageStartupMessages(library(magrittr))
if (rlang::is_true(getOption("knitr.in.progress"))) {
  params_ <- scdrake::scdrake_list(params)
}
drake_cache_dir <- params_$drake_cache_dir

drake::loadd(
  config_main, config_int_clustering,
  selected_markers_int_plots_final,
  dimred_plots_clustering_files, dimred_plots_clustering_files_out,
  dimred_plots_clustering_united_files, dimred_plots_clustering_united_files_out,
  cluster_graph_louvain_clustree_file, cluster_graph_leiden_clustree_file,
  cluster_sc3_clustree_file, cluster_sc3_cluster_stability_plots_file,
  cluster_kmeans_kbest_k, cluster_kmeans_k_clustree_file, cluster_kmeans_kbest_gaps_plot_file,
  dimred_plots_other_vars_files, dimred_plots_other_vars_files_out,
  selected_markers_int_plots_files, selected_markers_int_plots_files_out,
  dimred_plots_cell_annotation_files, dimred_plots_cell_annotation_files_out,
  cell_annotation_diagnostic_plots, cell_annotation_diagnostic_plots_files,
  dimred_plots_cell_annotation_files, dimred_plots_cell_annotation_files,

  path = drake_cache_dir
)

cfg <- config_int_clustering

if (!rlang::is_null(selected_markers_int_plots_files)) {
  selected_markers_int_plots_files <- dplyr::filter(
    selected_markers_int_plots_files,
    name == cfg$INTEGRATION_FINAL_METHOD,
    hvg_rm_cc_genes == cfg$INTEGRATION_FINAL_METHOD_RM_CC
  )
}

report_html_file <- cfg$INT_CLUSTERING_REPORT_HTML_FILE

any_clustering_enabled <- any(
  cfg$CLUSTER_GRAPH_LOUVAIN_ENABLED, cfg$CLUSTER_GRAPH_WALKTRAP_ENABLED, cfg$CLUSTER_GRAPH_LEIDEN_ENABLED,
  cfg$CLUSTER_KMEANS_K_ENABLED, cfg$CLUSTER_KMEANS_KBEST_ENABLED,
  cfg$CLUSTER_SC3_ENABLED
)

---

---

Input data overview

Just to review data from the preceding pipeline step (01 - integration).

int_method_desc <- scdrake::get_int_method_description(cfg$INTEGRATION_FINAL_METHOD)
scdrake::catg0("The chosen integration method was '{cfg$INTEGRATION_FINAL_METHOD}': {int_method_desc$header}\n\n")
if (cfg$INTEGRATION_FINAL_METHOD_RM_CC) {
  cat("\n\nCell cycle related genes were removed prior to HVG selection.\n\n")
}

Show integration method details ▾

r int_method_desc$fn_link

r int_method_desc$description

cat(drake::readd(sce_int_final_info, path = drake_cache_dir)$str)

---

if (any_clustering_enabled) {
  cat(knitr::knit_child(here::here("Rmd/common/clustering/clustering.Rmd"), quiet = TRUE))
  cat("\n\n#\n\n***\n\n")
}

if (!is.null(cfg$INT_CLUSTERING_REPORT_DIMRED_PLOTS_OTHER)) {
  res <- scdrake::generate_dimred_plots_section(
    dimred_plots_other_vars_files = dimred_plots_other_vars_files,
    selected_markers_plots_files = selected_markers_int_plots_files,
    dimred_plots_rel_start = fs::path_dir(report_html_file),
    selected_markers_files_rel_start = fs::path_dir(report_html_file),
    main_header = "Dimensionality reduction plots"
  )

  cat("\n\n#\n\n***\n\n")
}

if (!is.null(cfg$CELL_ANNOTATION_SOURCES)) {
  cell_annotation_text <- str_space(
    "We used the [SingleR](https://bioconductor.org/packages/3.15/bioc/html/SingleR.html) package to predict cell types in the dataset.",
    "Given a reference dataset of samples (single-cell or bulk) with known labels, `SinglerR` assigns those labels to",
    "new cells from a test dataset based on similarities in their expression profiles.",
    "You can find more information in the [SingleR book](https://bioconductor.org/books/3.15/SingleRBook/).\n\n",
    "The used references are shown below in the tabs. Each have several diagnostic plots:\n\n",
    "- Score heatmaps show distribution of predicted cell types in computed clusters (if any), along with per-cell annotation scores\n",
    "- Marker heatmaps show genes that are markers for a given cell type in both the reference and current datasets,",
    "i.e. those markers have driven the decision to label cells by the chosen cell type\n",
    "- Delta scores show poor-quality or ambiguous assignments based on the per-cell 'delta', i.e., the difference between",
    "the score for the assigned label and the median across all labels for each cell.",
    "See [OSCA](https://bioconductor.org/books/3.15/SingleRBook/annotation-diagnostics.html#based-on-the-deltas-across-cells) for more details"
  )

  res <- scdrake::generate_cell_annotation_plots_section(
    dimred_plots_cell_annotation_files = dimred_plots_cell_annotation_files,
    cell_annotation_diagnostic_plots = cell_annotation_diagnostic_plots,
    dimred_plots_rel_start = fs::path_dir(report_html_file),
    cell_annotation_diagnostic_plots_rel_start = fs::path_dir(report_html_file),
    main_header = "Cell annotation",
    text = cell_annotation_text
  )

  cat("\n\n#\n\n***\n\n")
}

Show input parameters

---

Main config

print(config_main)

---

Clustering config

print(cfg)

---

bioinfocz/scdrake documentation built on Jan. 29, 2024, 10:24 a.m.