biopipen 0.23.7__py3-none-any.whl → 0.24.0__py3-none-any.whl

biopipen/scripts/scrna/SeuratSubClustering.R

@@ -1,4 +1,5 @@
 source("{{biopipen_dir}}/utils/misc.R")
+source("{{biopipen_dir}}/utils/caching.R")
 
 library(Seurat)
 library(future)
@@ -33,40 +34,10 @@ envs$FindNeighbors <- .expand_dims(envs$FindNeighbors)
 log_info("Reading Seurat object ...")
 srtobj <- readRDS(srtfile)
 
-if (isTRUE(envs$cache)) {
-    envs$cache <- joboutdir
-}
-
-if (is.character(envs$cache) && nchar(envs$cache) > 0) {
-    log_info("Obtainning the signature ...")
-    envs2 <- envs
-    envs2$ncores <- NULL
-    sig <- c(
-        capture.output(str(srtobj)),
-        "\n\n-------------------\n\n",
-        capture.output(str(envs2)),
-        "\n"
-    )
-    digested_sig <- digest::digest(sig, algo = "md5")
-    cached_file <- file.path(envs$cache, paste0(digested_sig, ".cached.RDS"))
-    if (file.exists(cached_file)) {
-        log_info("Using cached results {cached_file}")
-        # copy cached file to rdsfile
-        file.copy(cached_file, rdsfile, copy.date = TRUE)
-        quit()
-    } else {
-        log_info("Cached results not found.")
-        log_info("- Current signature: {digested_sig}")
-        # print(sig)
-        # sigfiles <- Sys.glob(file.path(envs$cache, "*.signature.txt"))
-        # for (sigfile in sigfiles) {
-        #     log_info("- Found cached signature file: {sigfile}")
-        #     cached_sig <- readLines(sigfile)
-        #     log_info("- Cached signature:")
-        #     print(cached_sig)
-        # }
-        writeLines(sig, file.path(envs$cache, paste0(digested_sig, ".signature.txt")))
-    }
+if (isTRUE(envs$cache)) { envs$cache <- joboutdir }
+if (length(envs$cache) > 1) {
+    log_warn("Multiple cache directories (envs.cache) detected, using the first one.")
+    envs$cache <- envs$cache[1]
 }
 
 if (!is.null(envs$mutaters) && length(envs$mutaters) > 0) {
@@ -102,30 +73,66 @@ for (key in names(envs$cases)) {
     }
 
     log_info("- Subsetting ...")
-    sobj <- srtobj %>% filter(!!parse_expr(case$subset))
-
-    log_info("- Running RunUMAP ...")
-    umap_args <- list_setdefault(
-        case$RunUMAP,
-        object = sobj,
-        dims = 1:30,
-        reduction = sobj@misc$integrated_new_reduction %||% "pca"
-    )
-    umap_args$dims <- 1:min(max(umap_args$dims), ncol(sobj) - 1)
-    sobj <- do_call(RunUMAP, umap_args)
-
-    log_info("- Running FindNeighbors ...")
-    case$FindNeighbors$object <- sobj
-    if (is.null(case$FindNeighbors$reduction)) {
-        case$FindNeighbors$reduction <- sobj@misc$integrated_new_reduction %||% "pca"
+    sobj <- tryCatch({
+        srtobj %>% filter(!!parse_expr(case$subset))
+    }, error = function(e) {
+        stop(paste0("  Error in subset: ", e$message))
+    })
+    sobj_sig <- capture.output(str(sobj))
+    dig_sig <- digest::digest(sobj_sig, algo = "md5")
+    dig_sig <- substr(dig_sig, 1, 8)
+    cache_dir <- NULL
+    if (is.character(envs$cache)) {
+        cache_dir <- file.path(envs$cache, paste0(dig_sig, ".seurat_cache"))
+        dir.create(cache_dir, recursive = TRUE, showWarnings = FALSE)
+        writeLines(sobj_sig, file.path(cache_dir, "signature.txt"))
     }
-    sobj <- do_call(FindNeighbors, case$FindNeighbors)
 
-    log_info("- Running FindClusters ...")
-    if (is.null(case$FindClusters$random.seed)) {
-        case$FindClusters$random.seed <- 8525
+    cached <- get_cached(case$RunUMAP, "RunUMAP", cache_dir)
+    reduc_name <- case$RunUMAP$reduction.name %||% "umap"
+    if (is.null(cached$data)) {
+        log_info("- Running RunUMAP ...")
+        umap_args <- list_setdefault(
+            case$RunUMAP,
+            object = sobj,
+            dims = 1:30,
+            reduction = sobj@misc$integrated_new_reduction %||% "pca"
+        )
+        ncells <- ncol(sobj)
+        umap_args$dims <- 1:min(max(umap_args$dims), ncells - 1)
+        umap_method <- case$RunUMAP$umap.method %||% "uwot"
+        if (umap_method == "uwot" && is.null(case$RunUMAP$n.neighbors)) {
+            # https://github.com/satijalab/seurat/issues/4312
+            umap_args$n.neighbors <- min(ncells - 1, 30)
+        }
+        sobj <- do_call(RunUMAP, umap_args)
+        cached$data <- list(reduc = sobj@reductions[[reduc_name]], commands = sobj@commands)
+        save_to_cache(cached, "RunUMAP", cache_dir)
+    } else {
+        log_info("- Loading cached RunUMAP ...")
+        sobj@reductions[[reduc_name]] <- cached$data$reduc
+        sobj@commands <- cached$data$commands
+    }
+    reduc <- cached$data$reduc
+
+    cached <- get_cached(case$FindNeighbors, "FindNeighbors", cache_dir)
+    if (is.null(cached$data)) {
+        log_info("- Running FindNeighbors ...")
+        case$FindNeighbors$object <- sobj
+        if (is.null(case$FindNeighbors$reduction)) {
+            case$FindNeighbors$reduction <- sobj@misc$integrated_new_reduction %||% "pca"
+        }
+        sobj <- do_call(FindNeighbors, case$FindNeighbors)
+        cached$data <- list(graphs = sobj@graphs, commands = sobj@commands)
+        save_to_cache(cached, "FindNeighbors", cache_dir)
+    } else {
+        log_info("- Loading cached FindNeighbors ...")
+        sobj@graphs <- cached$data$graphs
+        sobj@commands <- cached$data$commands
     }
-    resolution <- case$FindClusters$resolution
+
+    case$FindClusters$random.seed <- case$FindClusters$random.seed %||% 8525
+    resolution <- case$FindClusters$resolution %||% 0.8
     if (is.character(resolution)) {
         if (grepl(",", resolution)) {
             resolution <- as.numeric(trimws(unlist(strsplit(resolution, ","))))
@@ -133,53 +140,30 @@ for (key in names(envs$cases)) {
             resolution <- as.numeric(resolution)
         }
     }
-    if (is.null(resolution) || length(resolution) == 1) {
-        case$FindClusters$resolution <- resolution
-        case$FindClusters$object <- sobj
-        sobj <- do_call(FindClusters, case$FindClusters)
-        levels(sobj$seurat_clusters) <- paste0("s", as.numeric(levels(sobj$seurat_clusters)) + 1)
-        Idents(sobj) <- "seurat_clusters"
-        sobj[[key]] <- sobj$seurat_clusters
-        ident_table <- table(sobj[[key]])
-        log_info("- Found {length(ident_table)} clusters:")
-        print(ident_table)
-        cat("\n")
-
-        log_info("- Updating meta.data with subclusters...")
-        srtobj <- AddMetaData(srtobj, metadata = sobj@meta.data[, key, drop = FALSE])
-        srtobj[[paste0("sub_umap_", key)]] <- sobj@reductions$umap
-    } else {
-        log_info("- Multiple resolutions detected ...")
-        log_info("")
-        metadata <- NULL
-        for (res in resolution) {
-            findclusters_args <- case$FindClusters
-            findclusters_args$resolution <- res
-            findclusters_args$object <- sobj
-            sobj1 <- do_call(FindClusters, findclusters_args)
-            res_key <- paste0(key, "_", res)
+    for (res in resolution) {
+        case$FindClusters$resolution <- res
+        cached <- get_cached(case$FindClusters, paste0("FindClusters_", res), cache_dir)
+        res_key <- paste0("seurat_clusters_", res)
+        if (is.null(cached$data)) {
+            log_info("- Running FindClusters at resolution: {res} ...")
+            case$FindClusters$object <- sobj
+            sobj1 <- do_call(FindClusters, case$FindClusters)
             levels(sobj1$seurat_clusters) <- paste0("s", as.numeric(levels(sobj1$seurat_clusters)) + 1)
-            Idents(sobj1) <- "seurat_clusters"
             sobj1[[res_key]] <- sobj1$seurat_clusters
-            ident_table <- table(sobj1[[res_key]])
-            log_info("- Found {length(ident_table)} at resolution: {res}:")
-            print(ident_table)
-            cat("\n")
-
-            log_info("- Updating meta.data with subclusters...")
-            metadata <- sobj1@meta.data[, res_key, drop = FALSE]
-            srtobj <- AddMetaData(srtobj, metadata = metadata)
-            srtobj[[paste0("sub_umap_", res_key)]] <- sobj1@reductions$umap
+            cached$data <- sobj1@meta.data[, res_key, drop = FALSE]
+            save_to_cache(cached, paste0("FindClusters_", res), cache_dir)
+        } else {
+            log_info("- Using cached FindClusters at resolution: {res} ...")
         }
-        srtobj <- AddMetaData(srtobj, metadata = metadata, col.name = key)
-        srtobj[[paste0("sub_umap_", key)]] <- sobj1@reductions$umap
+        ident_table <- table(cached$data[[res_key]])
+        log_info("  Found {length(ident_table)} clusters")
+        print(ident_table)
+        cat("\n")
     }
+    log_info("- Updating meta.data with subclusters...")
+    srtobj <- AddMetaData(srtobj, metadata = cached$data, col.name = key)
+    srtobj[[paste0("sub_umap_", key)]] <- reduc
 }
 
 log_info("Saving results ...")
 saveRDS(srtobj, file = rdsfile)
-
-if (is.character(envs$cache) && nchar(envs$cache) > 0) {
-    log_info("Caching results to {cached_file} ...")
-    invisible(file.copy(rdsfile, cached_file, overwrite = TRUE))
-}

biopipen/scripts/scrna_metabolic_landscape/MetabolicFeaturesIntraSubset.R

@@ -54,7 +54,7 @@ do_one_comparison <- function(
     subset_prefix,
     groupname
 ) {
-    print(paste("  Design:", compname, "(", case, ",", control, ")"))
+    log_info(paste("  Design: {compname} ({case}, {control})"))
     case_code = paste0("subset(obj, subset = ", subset_col, " == '", case, "')")
     case_obj = tryCatch({
         eval(parse(text = case_code))
@@ -62,7 +62,7 @@ do_one_comparison <- function(
         NULL
     })
     if (is.null(case_obj)) {
-        print("          Skip (not enough cells in case)")
+        log_warn("          Skip (not enough cells in case)")
         return (NULL)
     }
     control_code = paste0("subset(obj, subset = ", subset_col, " == '", control, "')")
@@ -72,7 +72,7 @@ do_one_comparison <- function(
         NULL
     })
     if (is.null(control_obj)) {
-        print("          Skip (not enough cells in control)")
+        log_warn("          Skip (not enough cells in control)")
         add_report(
             list(kind = "error", content = "Not enough cells in control"),
             h1 = groupname,
@@ -86,7 +86,7 @@ do_one_comparison <- function(
     odir = file.path(groupdir, paste0(subset_prefix, compname))
     dir.create(odir, showWarnings = FALSE)
     if (ncol(exprs_case) < 3 || ncol(exprs_control) < 3) {
-        print("          Skip (not enough cells)")
+        log_warn("          Skip (not enough cells)")
         add_report(
             list(kind = "error", content = "Not enough cells"),
             h1 = groupname,
@@ -131,7 +131,7 @@ do_one_comparison <- function(
 }
 
 do_one_group <- function(group) {
-    print(paste("- Group:", group, "..."))
+    log_info("- Group: {group} ...")
 
     genes = intersect(metabolics, rownames(sobj))
     group_code = paste0(

biopipen/scripts/scrna_metabolic_landscape/MetabolicPathwayActivity.R

@@ -71,7 +71,7 @@ num_of_pathways <- function(gmtfile, overlapgenes) {
 }
 
 do_one_subset <- function(s, subset_col, subset_prefix) {
-    print(paste0("  Processing subset: ", s, "..."))
+    log_info("  Processing subset: {s} ...")
     if (is.null(s)) {
         subset_dir <- file.path(outdir, "ALL")
         dir.create(subset_dir, showWarnings = FALSE)
@@ -118,7 +118,7 @@ do_one_subset <- function(s, subset_col, subset_prefix) {
 
     for (pi in seq_along(pathway_names)) {
         p <- pathway_names[pi]
-        print(paste0("  * Pathway (", pi, "): ", p, "..."))
+        log_info("  * Pathway ({pi}): {p} ...")
         genes <- pathways[[p]]
         genes_comm <- intersect(genes, rownames(subset_obj))
         genes_expressed <- names(rowSums(subset_obj)[rowSums(subset_obj) > 0])
@@ -312,7 +312,7 @@ do_one_subset <- function(s, subset_col, subset_prefix) {
 }
 
 do_one_subset_col <- function(subset_col, subset_prefix) {
-    print(paste0("- Handling subset column: ", subset_col, " ..."))
+    log_info("- Handling subset column: {subset_col} ...")
     if (is.null(subset_col)) {
         do_one_subset(NULL, subset_col = NULL, subset_prefix = NULL)
     } else {

biopipen/scripts/tcr/Immunarch-basic.R

@@ -2,9 +2,8 @@
 # immfile, outdir, mutaters, immdata, n_samples
 
 log_info("")
-log_info("#####################################")
-log_info("# Basic analysis                    #")
-log_info("#####################################")
+log_info("# Basic analysis")
+log_info("-----------------------------------")
 
 volumes = {{envs.volumes | r}}
 lens = {{envs.lens | r}}

biopipen/scripts/tcr/Immunarch-clonality.R

@@ -2,9 +2,8 @@
 # immfile, outdir, mutaters, immdata, n_samples
 
 log_info("")
-log_info("#####################################")
-log_info("# Clonality analysis                #")
-log_info("#####################################")
+log_info("# Clonality analysis")
+log_info("-----------------------------------")
 
 top_clones = {{envs.top_clones | r}}
 rare_clones = {{envs.rare_clones | r}}

biopipen/scripts/tcr/Immunarch-diversity.R

@@ -1,30 +1,34 @@
 # Diversity estimation
+source("{{biopipen_dir}}/scripts/tcr/immunarch-patched.R")
 # https://immunarch.com/articles/web_only/v6_diversity.html
 
 log_info("")
-log_info("#####################################")
-log_info("# Diversity estimation              #")
-log_info("#####################################")
+log_info("# Diversity estimation")
+log_info("-----------------------------------")
 
 # Other variables are loaded in the parent template
 # immdata is already loaded, meta is mutated
-div_method = {{envs.divs.method | r}}
-div_by = {{envs.divs.by | r}}
-div_order = {{envs.divs.order | r}}
-div_args = {{envs.divs.args | r: todot="-"}}
-div_test = {{envs.divs.test | r}}
-div_cases = {{envs.divs.cases | r: todot="-"}}
-div_devpars = {{envs.divs.devpars | r}}
-div_separate_by = {{envs.divs.separate_by | r}}
-div_split_by = {{envs.divs.split_by | r}}
-div_split_order = {{envs.divs.split_order | r}}
-div_align_x = {{envs.divs.align_x | r}}
-div_align_y = {{envs.divs.align_y | r}}
-div_subset = {{envs.divs.subset | r}}
-div_log = {{envs.divs.log | r}}
-div_ncol = {{envs.divs.ncol | r}}
-div_ymin = {{envs.divs.ymin | r}}
-div_ymax = {{envs.divs.ymax | r}}
+div_method = {{envs.divs.method | default: "gini" | r}}
+div_by = {{envs.divs.by | default: None | r}}
+div_plot_type = {{envs.divs.plot_type | default: "bar" | r}}
+div_order = {{envs.divs.order | default: [] | r}}
+div_args = {{envs.divs.args | default: {} | r: todot="-"}}
+div_test = {{envs.divs.test | default: None | r}}
+div_cases = {{envs.divs.cases | default: {} | r: todot="-"}}
+div_devpars = {{envs.divs.devpars | default: None | r}}
+div_separate_by = {{envs.divs.separate_by | default: None | r}}
+div_split_by = {{envs.divs.split_by | default: None | r}}
+div_split_order = {{envs.divs.split_order | default: None | r}}
+div_align_x = {{envs.divs.align_x | default: False | r}}
+div_align_y = {{envs.divs.align_y | default: False | r}}
+div_subset = {{envs.divs.subset | default: None | r}}
+div_log = {{envs.divs.log | default: False | r}}
+div_ncol = {{envs.divs.ncol | default: 2 | r}}
+div_ymin = {{envs.divs.ymin | default: None | r}}
+div_ymax = {{envs.divs.ymax | default: None | r}}
+
+div_test = div_test %||% list(method = "none", padjust = "none")
+div_devpars = div_devpars %||% list(res = 100, width = 800, height = 800)
 
 div_dir = file.path(outdir, "diversity")
 dir.create(div_dir, showWarnings = FALSE)
@@ -38,6 +42,7 @@ update_case = function(case, name) {
     if (!is.null(case$by) && nchar(case$by) > 0) {
         case$by = unlist(strsplit(case$by, ",")) %>% trimws()
     }
+    case$plot_type <- case$plot_type %||% div_plot_type
     case$order <- case$order %||% div_order
     case$args <- case$args %||% div_args
     for (name in names(case$args)) {
@@ -85,23 +90,6 @@ update_case = function(case, name) {
     return (case)
 }
 
-# See https://github.com/immunomind/immunarch/pull/341
-vis.immunr_gini <- function(.data, .by = NA, .meta = NA,
-                            .errorbars = c(0.025, 0.975), .errorbars.off = FALSE,
-                            .points = TRUE, .test = TRUE, .signif.label.size = 3.5, ...) {
-  # repDiversity(..., .method = "gini") generates a matrix
-  .data = data.frame(Sample = rownames(.data), Value = .data[, 1])
-  vis_bar(
-    .data = .data, .by = .by, .meta = .meta,
-    .errorbars = .errorbars, .errorbars.off = .errorbars.off, .stack = FALSE,
-    .points = .points, .test = .test, .signif.label.size = .signif.label.size,
-    .defgroupby = "Sample", .grouping.var = "Group",
-    .labs = c(NA, "Gini coefficient"),
-    .title = "Gini coefficient", .subtitle = "Sample diversity estimation using the Gini coefficient",
-    .legend = NA, .leg.title = NA
-  )
-}
-
 if (is.null(div_cases) || length(div_cases) == 0) {
     if (is.null(div_method) || length(div_method) == 0 || nchar(div_method) == 0) {
         stop("No method is specified for diversity estimation")
@@ -176,6 +164,15 @@ run_general = function(casename, d, case, ddir, value_col = "Value") {
         col.names = TRUE
     )
 
+    .meta_vals <- function(meta, cols) {
+        if (length(cols) == 1) {
+            return (meta[[cols]])
+        }
+
+        vlist = lapply(cols, function(.x) meta[[.x]])
+        do.call(function(...) paste(..., sep = "; "), vlist)
+    }
+
     # plot
     #  by, order, separate_by, align_y
     n_seps = 1
@@ -189,11 +186,19 @@ run_general = function(casename, d, case, ddir, value_col = "Value") {
                 metas = metas[intersect(case$split_order, names(metas))]
             }
             ps = lapply(metas, function(meta) {
-                .test = length(unique(meta[[case$by]])) > 1
-                p = vis(filter_div(div, meta$Sample), .by = case$by, .meta = meta, .test = .test)
+                .test = length(unique(.meta_vals(meta, case$by))) > 1
+                p = vis(
+                    filter_div(div, meta$Sample),
+                    .by = case$by,
+                    .meta = meta,
+                    .test = .test,
+                    .plot.type = case$plot_type
+                )
                 p = p + xlab(paste0(case$separate_by, ": ", meta[[case$separate_by]][1], ")"))
                 if (!is.null(case$order) && length(case$order) > 0) {
-                    p = p + scale_x_discrete(limits = intersect(case$order, unique(meta[[case$by]])))
+                    p = p + scale_x_discrete(
+                        limits = intersect(case$order, unique(.meta_vals(meta, case$by)))
+                    )
                 }
                 if (!is.null(case$ymin) && !is.null(case$ymax)) {
                     p = p + ylim(c(case$ymin, case$ymax))
@@ -217,10 +222,18 @@ run_general = function(casename, d, case, ddir, value_col = "Value") {
             }
             .i = 0
             ps = lapply(metas, function(meta) {
-                nby = length(unique(meta[[case$by]]))
-                p = vis(filter_div(div, meta$Sample), .by = case$by, .meta = meta, .test = nby > 1)
+                nby = length(unique(.meta_vals(meta, case$by)))
+                p = vis(
+                    filter_div(div, meta$Sample),
+                    .by = case$by,
+                    .meta = meta,
+                    .test = nby > 1,
+                    .plot.type = case$plot_type
+                )
                 if (!is.null(case$order) && length(case$order) > 0) {
-                    p = p + scale_x_discrete(limits = intersect(case$order, unique(meta[[case$by]])))
+                    p = p + scale_x_discrete(
+                        limits = intersect(case$order, unique(.meta_vals(meta, case$by)))
+                    )
                 }
                 p = p + xlab(meta[[case$split_by]][1]) + theme(
                     axis.text.x = element_blank(),
@@ -253,10 +266,10 @@ run_general = function(casename, d, case, ddir, value_col = "Value") {
             plots = lapply(ps, function(x) x$p + ylim(c(ymin, ymax)))
             p = wrap_plots(plots, widths = widths, guides = "collect")
         } else {
-            .test = length(unique(d$meta[[case$by]])) > 1
-            p = vis(div, .by = case$by, .meta = d$meta, .test = .test)
+            .test = length(unique(.meta_vals(d$meta, case$by))) > 1
+            p = vis(div, .by = case$by, .meta = d$meta, .test = .test, .plot.type = case$plot_type)
             if (!is.null(case$order) && length(case$order) > 0) {
-                p = p + scale_x_discrete(limits = intersect(case$order, unique(d$meta[[case$by]])))
+                p = p + scale_x_discrete(limits = intersect(case$order, unique(.meta_vals(d$meta, case$by))))
             }
         }
     } else if (!is.null(case$separate_by)) {
@@ -333,7 +346,9 @@ run_general = function(casename, d, case, ddir, value_col = "Value") {
     } else {
         p = vis(div)
         if (!is.null(case$order) && length(case$order) > 0) {
-            p = p + scale_x_discrete(limits = intersect(case$order, unique(d$meta[[case$by]])))
+            p = p + scale_x_discrete(
+                limits = intersect(case$order, unique(.meta_vals(d$meta, case$by)))
+            )
         }
     }
 
@@ -351,7 +366,7 @@ run_general = function(casename, d, case, ddir, value_col = "Value") {
     }
     if (is.null(width)) {
         if (!is.null(case$by) && length(case$by) > 0) {
-            width = 200 * length(unique(d$meta[[case$by]])) + 120
+            width = 200 * length(unique(.meta_vals(d$meta, case$by))) + 120
         } else {
             width = 100 * length(unique(d$meta$Sample)) + 120
         }
@@ -400,7 +415,11 @@ run_general = function(casename, d, case, ddir, value_col = "Value") {
                         "where all values are the same (for example, where everyone has ",
                         "the same income). A Gini coefficient of one (or 100 percents ) ",
                         "expresses maximal inequality among values (for example where only ",
-                        "one person has all the income).")
+                        "one person has all the income)."),
+                    d50 = paste0(
+                        "the D50 index. ",
+                        "It is the number of types that are needed to cover 50% of the total
+                        abundance.")
                 )
             )
         ),
@@ -705,6 +724,8 @@ run_div_case = function(casename) {
             run_general(casename, d, case, ddir)
         } else if (case$method == "gini") {
             run_general(casename, d, case, ddir, "V1")
+        } else if (case$method == "d50") {
+            run_general(casename, d, case, ddir, "Clones")
         } else {
             stop(paste0("Unknown diversity method: ", case$method))
         }

biopipen/scripts/tcr/Immunarch-geneusage.R

@@ -2,9 +2,8 @@
 # immfile, outdir, mutaters, immdata, n_samples
 
 log_info("")
-log_info("#####################################")
-log_info("# Gene usage analysis               #")
-log_info("#####################################")
+log_info("# Gene usage analysis")
+log_info("-----------------------------------")
 
 gene_usages = {{ envs.gene_usages | r: todot="-" }}
 

biopipen/scripts/tcr/Immunarch-kmer.R

@@ -2,9 +2,8 @@
 # immfile, outdir, mutaters, immdata, n_samples
 
 log_info("")
-log_info("#####################################")
-log_info("# K-mer analysis                    #")
-log_info("#####################################")
+log_info("# K-mer analysis")
+log_info("-----------------------------------")
 
 kmers = {{ envs.kmers | r: todot="-" }}
 

biopipen/scripts/tcr/Immunarch-overlap.R

@@ -2,9 +2,8 @@
 # immfile, outdir, mutaters, immdata, n_samples
 
 log_info("")
-log_info("#####################################")
-log_info("# Overlap analysis                  #")
-log_info("#####################################")
+log_info("# Overlap analysis")
+log_info("-----------------------------------")
 
 overlaps = {{ envs.overlaps | r: todot="-" }}
 

biopipen/scripts/tcr/Immunarch-spectratyping.R

@@ -2,9 +2,8 @@
 # immfile, outdir, mutaters, immdata, n_samples
 
 log_info("")
-log_info("#####################################")
-log_info("# Spectratyping analysis            #")
-log_info("#####################################")
+log_info("# Spectratyping analysis")
+log_info("-----------------------------------")
 
 spects = {{ envs.spects | r }}
 

biopipen/scripts/tcr/Immunarch-tracking.R

@@ -1,7 +1,6 @@
 log_info("")
-log_info("#####################################")
-log_info("# Clonotype tracking                #")
-log_info("#####################################")
+log_info("# Clonotype tracking")
+log_info("-----------------------------------")
 
 trackings = {{ envs.trackings | r }}
 

biopipen/scripts/tcr/Immunarch-vjjunc.R

@@ -1,7 +1,6 @@
 log_info("")
-log_info("#####################################")
-log_info("# VJ Junction Circos Plots          #")
-log_info("#####################################")
+log_info("# VJ Junction Circos Plots")
+log_info("-----------------------------------")
 
 # Already required by immunarch
 library(circlize)

biopipen/scripts/tcr/Immunarch.R

@@ -34,7 +34,7 @@ log_info("Expanding immdata ...")
 exdata = expand_immdata(immdata)
 
 log_info("Loading metadata if provided ...")
-if (endsWith(metafile, ".rds") || endsWith(metafile, ".RDS")) {
+if (!is.null(metafile) && (endsWith(metafile, ".rds") || endsWith(metafile, ".RDS"))) {
     meta = readRDS(metafile)@meta.data
 } else if (!is.null(metafile) && nchar(metafile) > 0) {
     meta = read.table(metafile, sep = "\t", header = TRUE, row.names = 1)

biopipen/scripts/tcr/ImmunarchLoading.R

@@ -144,6 +144,7 @@ for (i in seq_len(nrow(metadata))) {
     # file.symlink(normalizePath(annofile), file.path(datadir, paste0(sample, ext)))
 }
 
+log_info("Loading TCR data ...")
 immdata = repLoad(datadir, .mode=mode)
 if (mode == "single") {
     data = immdata$data
@@ -178,6 +179,7 @@ immdata$prefix = prefix
 
 saveRDS(immdata, file=rdsfile)
 
+log_info("Saving cell-level data ...")
 exdata <- expand_immdata(immdata, cell_id = "Barcode") %>%
     distinct(Sample, Barcode, .keep_all = TRUE) %>%
     mutate(Barcode = glue(paste0(prefix, "{Barcode}"))) %>%

biopipen/scripts/tcr/TCRClustering.R

@@ -3,6 +3,7 @@
 # python = Sys.which({{envs.python | r}})
 # Sys.setenv(RETICULATE_PYTHON = python)
 # library(reticulate)
+source("{{biopipen_dir}}/utils/misc.R")
 source("{{biopipen_dir}}/utils/single_cell.R")
 
 library(immunarch)
@@ -97,7 +98,7 @@ clean_clustcr_output = function(clustcr_outfile, clustcr_input) {
 }
 
 run_clustcr = function() {
-    print(paste("Using tool:", "ClusTCR"))
+    log_info("Running ClusTCR ...")
     clustcr_dir = file.path(outdir, "ClusTCR_Output")
     dir.create(clustcr_dir, showWarnings = FALSE)
     clustcr_file = prepare_clustcr(clustcr_dir)
@@ -110,6 +111,7 @@ run_clustcr = function() {
     )
     print("Running:")
     print(clustcr_cmd)
+    log_debug("- Running command: {clustcr_cmd}")
     rc = system(clustcr_cmd)
     if (rc != 0) {
         quit(status=rc)
@@ -196,7 +198,7 @@ clean_giana_output = function(giana_outfile, giana_infile) {
 }
 
 run_giana = function() {
-    print(paste("Using tool:", "GIANA"))
+    log_info("Running GIANA ...")
     giana_srcdir = prepare_giana()
     giana_input = prepare_input()
     giana_outdir = file.path(outdir, "GIANA_Output")
@@ -228,6 +230,7 @@ run_giana = function() {
     )
     print("Running:")
     print(giana_cmd)
+    log_debug("- Running command: {giana_cmd}")
     rc = system(giana_cmd)
     if (rc != 0) {
         quit(status=rc)
@@ -276,4 +279,5 @@ if (tolower(tool) == "clustcr") {
     stop(paste("Unknown tool:", tool))
 }
 
+log_info("Saving results ...")
 attach_to_immdata(out)