biopipen 0.34.1__py3-none-any.whl → 0.34.3__py3-none-any.whl

biopipen/scripts/scrna/ScFGSEA.R

@@ -7,9 +7,9 @@ srtfile <- {{in.srtobj | r}}  # nolint
 outdir <- {{out.outdir | r}}  # nolint
 joboutdir <- {{job.outdir | r}}  # nolint
 mutaters <- {{envs.mutaters | r}}  # nolint
-group.by <- {{envs["group-by"] | r}}  # nolint
-ident.1 <- {{envs["ident-1"] | r}}  # nolint
-ident.2 <- {{envs["ident-2"] | r}}  # nolint
+group_by <- {{envs.group_by | default: envs["group-by"] | default: None | r}}  # nolint
+ident_1 <- {{envs.ident_1 | default: envs["ident-1"] | default: None | r}}  # nolint
+ident_2 <- {{envs.ident_2 | default: envs["ident-2"] | default: None | r}}  # nolint
 each <- {{envs.each | r}}  # nolint
 subset <- {{envs.subset | r}}  # nolint
 gmtfile <- {{envs.gmtfile | r}}  # nolint
@@ -18,8 +18,8 @@ top <- {{envs.top | r}}  # nolint
 minsize <- {{envs.minSize | default: envs.minsize | r}}  # nolint
 maxsize <- {{envs.maxSize | default: envs.maxsize | r}}  # nolint
 eps <- {{envs.eps | r}}  # nolint
-allpathway_plots_defaults <- {{envs.allpathway_plots_defaults | r}}  # nolint
-allpathway_plots <- {{envs.allpathway_plots | r}}  #
+alleach_plots_defaults <- {{envs.alleach_plots_defaults | r}}  # nolint
+alleach_plots <- {{envs.alleach_plots | r}}  #
 ncores <- {{envs.ncores | r}}  # nolint
 rest <- {{envs.rest | r: todot="-"}}  # nolint
 cases <- {{envs.cases | r: todot="-"}}  # nolint
@@ -27,8 +27,8 @@ cases <- {{envs.cases | r: todot="-"}}  # nolint
 log <- get_logger()
 reporter <- get_reporter()
 
-allpathway_plots <- lapply(allpathway_plots, function(x) {
-    list_update(allpathway_plots_defaults, x)
+alleach_plots <- lapply(alleach_plots, function(x) {
+    list_update(alleach_plots_defaults, x)
 })
 
 log$info("Reading Seurat object ...")
@@ -43,9 +43,9 @@ if (!is.null(mutaters) && length(mutaters) > 0) {
 }
 
 defaults <- list(
-    group.by = group.by,
-    ident.1 = ident.1,
-    ident.2 = ident.2,
+    group_by = group_by,
+    ident_1 = ident_1,
+    ident_2 = ident_2,
     each = each,
     subset = subset,
     gmtfile = gmtfile,
@@ -54,8 +54,8 @@ defaults <- list(
     minsize = minsize,
     maxsize = maxsize,
     eps = eps,
-    allpathway_plots_defaults = allpathway_plots_defaults,
-    allpathway_plots = allpathway_plots,
+    alleach_plots_defaults = alleach_plots_defaults,
+    alleach_plots = alleach_plots,
     ncores = ncores,
     rest = rest
 )
@@ -63,11 +63,11 @@ defaults <- list(
 expand_each <- function(name, case) {
     outcases <- list()
 
-    case$group.by <- case$group.by %||% "Identity"
+    case$group_by <- case$group_by %||% "Identity"
 
     if (is.null(case$each) || is.na(case$each) || nchar(case$each) == 0 || isFALSE(each)) {
-        if (length(case$allpathway_plots) > 0) {
-            stop("Cannot perform `allpathway_plots` without `each` defined.")
+        if (length(case$alleach_plots) > 0) {
+            stop("Cannot perform `alleach_plots` without `each` defined.")
         }
 
         outcases[[name]] <- case
@@ -93,8 +93,8 @@ expand_each <- function(name, case) {
             newcase$each_name <- case$each
             newcase$each <- each
 
-            newcase$allpathway_plots_defaults <- NULL
-            newcase$allpathway_plots <- NULL
+            newcase$alleach_plots_defaults <- NULL
+            newcase$alleach_plots <- NULL
 
             if (!is.null(case$subset)) {
                 newcase$subset <- paste0(case$subset, " & ", bQuote(case$each), " == '", each, "'")
@@ -105,13 +105,13 @@ expand_each <- function(name, case) {
             outcases[[newname]] <- newcase
         }
 
-        if (length(case$allpathway_plots) > 0) {
+        if (length(case$alleach_plots) > 0) {
             newcase <- case
 
             newcase$gseas <- list()
-            newcase$allpathway_plots <- lapply(
-                newcase$allpathway_plots,
-                function(x) { list_update(newcase$allpathway_plots_defaults, x) }
+            newcase$alleach_plots <- lapply(
+                newcase$alleach_plots,
+                function(x) { list_update(newcase$alleach_plots_defaults, x) }
             )
 
             outcases[[paste0(name, " (all ", case$each,")")]] <- newcase
@@ -154,8 +154,8 @@ do_case <- function(name) {
         }))
         gseas[[case$each]] <- factor(gseas[[case$each]], levels = each_levels)
 
-        for (plotname in names(case$allpathway_plots)) {
-            plotargs <- case$allpathway_plots[[plotname]]
+        for (plotname in names(case$alleach_plots)) {
+            plotargs <- case$alleach_plots[[plotname]]
             plotargs <- extract_vars(plotargs, "devpars")
             plotargs$gsea_results <- gseas
             plotargs$group_by <- case$each
@@ -182,12 +182,12 @@ do_case <- function(name) {
     allow_empty = !is.null(case$each)
     # prepare expression matrix
     log$info("  Preparing expression matrix...")
-    sobj <- ensure_sobj({ srtobj %>% filter(!is.na(!!sym(case$group.by))) }, allow_empty)
+    sobj <- ensure_sobj({ srtobj %>% filter(!is.na(!!sym(case$group_by))) }, allow_empty)
     if (is.null(sobj)) {
         reporter$add2(
             list(
                 kind = "error",
-                content = paste0("No cells with non-NA `", case$group.by, "` in the Seurat object.")
+                content = paste0("No cells with non-NA `", case$group_by, "` in the Seurat object.")
             ),
             hs = c(info$section, info$name)
         )
@@ -200,20 +200,20 @@ do_case <- function(name) {
             reporter$add2(
                 list(
                     kind = "error",
-                    content = paste0("No cells with non-NA `", case$group.by, "` in the Seurat object.")
+                    content = paste0("No cells with non-NA `", case$group_by, "` in the Seurat object.")
                 ),
                 hs = c(info$section, info$name)
             )
             return(NULL)
         }
     }
-    if (!is.null(case$ident.2)) {
-        sobj <- ensure_sobj({ sobj %>% filter(!!sym(case$group.by) %in% c(case$ident.1, case$ident.2)) }, allow_empty)
+    if (!is.null(case$ident_2)) {
+        sobj <- ensure_sobj({ sobj %>% filter(!!sym(case$group_by) %in% c(case$ident_1, case$ident_2)) }, allow_empty)
         if (is.null(sobj)) {
             reporter$add2(
                 list(
                     kind = "error",
-                    content = paste0("No cells with non-NA `", case$group.by, "` in the Seurat object.")
+                    content = paste0("No cells with non-NA `", case$group_by, "` in the Seurat object.")
                 ),
                 hs = c(info$section, info$name)
             )
@@ -221,16 +221,16 @@ do_case <- function(name) {
         }
     }
 
-    allclasses <- sobj@meta.data[, case$group.by, drop = TRUE]
-    if (is.null(case$ident.2)) {
-        case$ident.2 <- "Other"
-        allclasses[allclasses != case$ident.1] <- "Other"
+    allclasses <- sobj@meta.data[, case$group_by, drop = TRUE]
+    if (is.null(case$ident_2)) {
+        case$ident_2 <- "Other"
+        allclasses[allclasses != case$ident_1] <- "Other"
     }
     exprs <- GetAssayData(sobj, layer = "data")
 
     # get preranks
     log$info("  Getting preranks...")
-    ranks <- RunGSEAPreRank(exprs, allclasses, case$ident.1, case$ident.2, case$method)
+    ranks <- RunGSEAPreRank(exprs, allclasses, case$ident_1, case$ident_2, case$method)
     write.table(
         as.data.frame(ranks),
         file.path(info$prefix, "fgsea.rank.txt"),
@@ -310,7 +310,7 @@ do_case <- function(name) {
 
     reporter$add2(
         list(
-            name = paste0("Table (", case$ident.1, " vs ", case$ident.2, ")"),
+            name = paste0("Table (", case$ident_1, " vs ", case$ident_2, ")"),
             contents = list(
                 list(kind = "descr", content = paste0(
                     "Showing top 50 pathways by padj in descending order. ",

biopipen/scripts/scrna/SeuratClusterStats-clustree.R

@@ -26,6 +26,22 @@ if (
     if (length(clustrees) == 0) {
         log$warn("- no case found, skipping ...")
     } else {
+        reporter$add(
+            list(
+                kind = "descr",
+                content = 'The clustree plots displays clustering results from the Seurat object across different
+                resolutions of the clustering algorithm
+                (<a target="_blank" href="https://satijalab.org/seurat/reference/findclusters">Seurat::FindClusters</a>).
+                Each node represents a cluster, with the resolution levels labeled along the vertical (y) axis.
+                The size of each node reflects the number of cells in that cluster. Edges connect clusters between
+                adjacent resolutions and indicate how cells transition between clusters as resolution increases.
+                The thickness of the edges corresponds to the proportion of shared cells (in_prop) between clusters,
+                where darker lines signify a higher overlap (up to 100%). The color of the edges indicates the actual
+                number of cells that transitioned between clusters.'
+            ),
+            h1 = "Clustree plots"
+        )
+
         reports <- list()
         for (name in names(clustrees)) {
             if (is.null(clustrees[[name]]$prefix)) {

biopipen/scripts/scrna/SeuratClusterStats-dimplots.R

@@ -40,7 +40,7 @@ do_one_dimplot = function(name) {
     reporter$add(
         list(
             kind = "descr",
-            content = paste0("Dimensionality reduction plot for ", case$group.by)
+            content = paste0("Dimensionality reduction plot for ", case$group_by)
         ),
         reporter$image(prefix, "pdf", FALSE),
         h1 = name

biopipen/scripts/scrna/SeuratClusterStats-features.R

@@ -64,11 +64,11 @@ do_one_features <- function(name) {
     log$info("- Case: {name}")
 
     case <- list_update(features_defaults, features[[name]])
-    case$descr <- case$descr %||% ""
     case <- extract_vars(
         case,
         "devpars", "more_formats", "save_code", "save_data", "order_by",
-        "subset", "features", "descr")
+        "subset", "features", "descr",
+        allow_nonexisting = TRUE)
 
     if (!is.null(subset)) {
         case$object <- srtobj %>% filter(!!parse_expr(subset))
@@ -77,6 +77,7 @@ do_one_features <- function(name) {
     }
 
     if (exists("order_by") && !is.null(order_by)) {
+        case$ident <- case$ident %||% GetIdentityColumn(case$object)
         if (length(order_by) < 2) {
             clusters <- case$object@meta.data %>%
                 group_by(!!sym(case$ident)) %>%
@@ -126,12 +127,34 @@ do_one_features <- function(name) {
         caching$save(info$prefix)
     }
     # add reports
-    if (!is.null(descr) && nchar(descr) > 0) {
-        reporter$add2(
-            list(kind = "descr", content = descr),
-            hs = c(info$section, info$name)
+    default_descr <- glue(
+        "The plot shows the distribution or pattern of the specified features ({paste(case$features %||% features, collapse = ', ')}) ",
+        "across cells",
+        "{if (!is.null(case$ident)) glue(', identified by \"{case$ident}\"') else ''}",
+        "{if (!is.null(case$group_by)) glue(', grouped by \"{case$group_by}\"') else ''}",
+        "{if (!is.null(case$split_by)) glue(', and split by \"{case$split_by}\"') else ''}. ",
+        "The plot type is '{case$plot_type}', ",
+        "{if (case$plot_type == 'dim') 'displaying the features on a dimensional reduction embedding' ",
+        " else if (case$plot_type == 'heatmap') 'arranged as a heatmap by rows_name and other grouping variables' ",
+        " else if (case$plot_type %in% c('violin', 'box', 'ridge')) 'showing the distribution of feature values by the grouping variables' ",
+        " else if (case$plot_type == 'cor') 'showing the correlation between features' ",
+        " else 'showing aggregated feature values by the grouping variables'}. ",
+        "{if (!is.null(case$facet_by)) glue('Plots are further faceted by \"{case$facet_by}\". ') else ''}",
+        "{if (case$plot_type == 'dim') glue('The reduction used is \"{if (!is.null(case$reduction)) case$reduction else DefaultDimReduc(case$object)}\"') else ''}",
+        "{if (case$plot_type == 'dim' && !is.null(case$graph)) glue(', with graph \"{case$graph}\" drawn to show cell neighbor edges') else ''}",
+        "{if (case$plot_type == 'dim' && !is.null(case$bg_cutoff) && case$bg_cutoff > 0) glue(', and a background cutoff of {case$bg_cutoff}') else ''}",
+        "{if (case$plot_type == 'dim') glue(', using dimensions {paste(case$dims %||% 1:2, collapse = \",\")}') else ''}"
+    )
+    if (!is.null(case$comparisons)) {
+        default_descr <- paste0(
+            default_descr,
+            "Statistical comparisons were performed between groups using '{case$pairwise_method %||% 'wilcox.test'}' method."
         )
     }
+    reporter$add2(
+        list(kind = "descr", content = descr %||% default_descr),
+        hs = c(info$section, info$name)
+    )
 
     if (save_data) {
         reporter$add2(

biopipen/scripts/scrna/SeuratClusterStats-stats.R

@@ -5,17 +5,26 @@ log$info("stats:")
 odir <- file.path(outdir, "stats")
 dir.create(odir, recursive=TRUE, showWarnings=FALSE)
 
+
+
 do_one_stats <- function(name) {
     log$info("- Case: {name}")
 
     case <- list_update(stats_defaults, stats[[name]])
-    extract_vars(case, "devpars", "more_formats", "save_code", "save_data", "subset")
+    case <- extract_vars(case, "devpars", "more_formats", "save_code", "save_data", "subset", "descr")
 
     if (!is.null(subset)) {
         case$object <- srtobj %>% filter(!!parse_expr(subset))
     } else {
         case$object <- srtobj
     }
+    ident <- case$ident %||% GetIdentityColumn(case$object)
+    groupings <- unique(c(case$group_by, case$rows_by, case$columns_by, case$pie_group_by, ident))
+    if (length(groupings) > 0) {
+        for (g in groupings) {
+            case$object <- filter(case$object, !is.na(!!sym(g)))
+        }
+    }
 
     info <- case_info(name, odir, is_dir = FALSE, create = TRUE)
     p <- do_call(gglogger::register(CellStatPlot), case)
@@ -27,6 +36,20 @@ do_one_stats <- function(name) {
             auto_data_setup = FALSE)
     }
 
+    frac <- case$frac %||% "none"
+    default_descr <- glue(
+        "The {case$plot_type} plot shows the distribution of cells across categories defined by '{ident}'",
+        "{if (!is.null(case$group_by)) glue(', grouped by {case$group_by}') else ''}",
+        "{if (!is.null(case$split_by)) glue(', and split by {case$split_by}') else ''}. ",
+        "The values represent ",
+        "{if (frac == 'none') 'the number of cells' else glue('the fraction of cells calculated by \"{frac}\"')}. "
+    )
+    if (!is.null(case$comparisons)) {
+        default_descr <- paste0(
+            default_descr,
+            "Statistical comparisons were performed between groups using '{case$pairwise_method %||% 'wilcox.test'}' method."
+        )
+    }
     if (save_data) {
         pdata <- attr(p, "data") %||% p$data
         if (!inherits(pdata, "data.frame") && !inherits(pdata, "matrix")) {
@@ -37,6 +60,10 @@ do_one_stats <- function(name) {
             list(
                 name = "Plot",
                 contents = list(
+                    list(
+                        kind = "descr",
+                        content = case$descr %||% default_descr
+                    ),
                     reporter$image(
                         info$prefix, more_formats, save_code, kind = "image")
                 )
@@ -60,6 +87,7 @@ do_one_stats <- function(name) {
         )
     } else {
         reporter$add2(
+            list(kind = "descr", content = case$descr %||% default_descr),
             reporter$image(info$prefix, more_formats, save_code, kind = "image"),
             hs = c(info$section, info$name)
         )

biopipen/scripts/scrna/SeuratClusterStats.R

@@ -3,6 +3,7 @@ library(rlang)
 library(dplyr)
 library(tidyr)
 library(tibble)
+library(glue)
 library(forcats)
 library(tidyseurat)
 library(gglogger)

biopipen/scripts/scrna/TopExpressingGenes.R

@@ -9,7 +9,7 @@ outdir <- {{out.outdir | r}}
 joboutdir <- {{job.outdir | r}}
 mutaters <- {{ envs.mutaters | r }}
 ident <- {{ envs.ident | r }}
-group.by <- {{ envs["group-by"] | r }}  # nolint
+group_by <- {{ envs.group_by | default: envs["group-by"] | default: None | r }}  # nolint
 each <- {{ envs.each | r }}
 dbs <- {{ envs.dbs | r }}
 n <- {{ envs.n | r }}
@@ -41,7 +41,7 @@ enrich_plots <- lapply(enrich_plots, function(x) {
 })
 defaults <- list(
     ident = ident,
-    group.by = group.by,
+    group_by = group_by,
     each = each,
     dbs = dbs,
     n = n,
@@ -171,17 +171,17 @@ run_case <- function(name) {
     } else {
         subobj <- srtobj
     }
-    case$group.by <- case$group.by %||% "Identity"
+    case$group_by <- case$group_by %||% "Identity"
     if (is.null(case$ident)) {
-        case$ident <- as.character(unique(subobj@meta.data[[case$group.by]]))
+        case$ident <- as.character(unique(subobj@meta.data[[case$group_by]]))
     }
     avgexpr <- AverageExpression(
         subobj,
-        group.by = case$group.by,
+        group_by = case$group_by,
         assays = assay
     )[[assay]]
     # https://github.com/satijalab/seurat/issues/7893
-    colnames(avgexpr) <- as.character(unique(subobj@meta.data[[case$group.by]]))
+    colnames(avgexpr) <- as.character(unique(subobj@meta.data[[case$group_by]]))
     avgexpr <- avgexpr[, case$ident, drop = FALSE]
 
     for (idt in case$ident) {

biopipen/scripts/scrna/celltypist-wrapper.py

@@ -29,6 +29,8 @@ if __name__ == "__main__":
         raise ValueError(
             f"Over clustering column '{over_clustering}' not found in AnnData object."
         )
+    if 'neighbors' in adata.uns and 'params' in adata.uns['neighbors']:
+        adata.uns['neighbors']['params'].setdefault('n_neighbors', 15)
 
     annotated = celltypist.annotate(
         adata,

biopipen/scripts/scrna_metabolic_landscape/MetabolicFeatures.R

@@ -98,7 +98,13 @@ do_comparison <- function(object, caseinfo, subset_by, subset_val, group_by, gro
     }
 
     classes <- as.character(object@meta.data[[group_by]])
-    classes[classes != group1] <- "_REST"
+    if (!group1 %in% classes) {
+        stop("Group '", group1, "' not found in '", group_by, "' column of the Seurat object.")
+    }
+    if (!is.null(group2) && !group2 %in% classes) {
+        stop("Group '", group2, "' not found in '", group_by, "' column of the Seurat object.")
+    }
+    classes[classes != group1] <- "Other"
     if (any(table(classes) < 5)) {
         msg <- paste0(
             "  ! skipped. Group has less than 5 cells: ",
@@ -266,8 +272,8 @@ do_subset <- function(object, caseinfo, subset_by, subset_val, group_by, compari
             rbind, lapply(
                 as.character(comparisons),
                 function(comparison) {
-                    if (grepl(",", comparison)) {
-                        group1 <- trimws(unlist(strsplit(comparison, ",")))
+                    if (grepl(":", comparison)) {
+                        group1 <- trimws(unlist(strsplit(comparison, ":")))
                         group2 <- group1[2]
                         group1 <- group1[1]
                     } else {

biopipen/scripts/scrna_metabolic_landscape/MetabolicPathwayActivity.R

@@ -315,8 +315,8 @@ do_subset <- function(
             plotargs$keep_empty <- TRUE
 
             p <- do_call(plotfn, plotargs)
-            devpars$width <- devpars$width %||% (attr(p, "width") * devpars$res) %||% 1000
-            devpars$height <- devpars$height %||% (attr(p, "height") * devpars$res) %||% 1000
+            devpars$width <- devpars$width %||% (attr(p, "width") * 2 * devpars$res) %||% 1000
+            devpars$height <- devpars$height %||% (attr(p, "height") * 2 * devpars$res) %||% 1000
         } else {  # heatmap
             minval <- min(dat)
             maxval <- max(dat)

biopipen/scripts/scrna_metabolic_landscape/MetabolicPathwayHeterogeneity.R

@@ -195,6 +195,7 @@ do_subset <- function(object, caseinfo, subset_by, subset_val, group_by, plots,
         plotprefix <- file.path(odir, slugify(plot))
         plotargs$devpars$width <- plotargs$devpars$width %||% (attr(p, "width") * plotargs$devpars$res) %||% 800
         plotargs$devpars$height <- plotargs$devpars$height %||% (attr(p, "height") * plotargs$devpars$res) %||% 600
+        plotargs$devpars$height <- max(plotargs$devpars$height, plotargs$devpars$width / 1.5)
         png(
             filename = paste0(plotprefix, ".png"),
             width = plotargs$devpars$width,

biopipen/scripts/tcr/GIANA/GIANA4.py

@@ -36,9 +36,6 @@ from sklearn.manifold import MDS
 import faiss
 from query import *
 try:
-    from Bio.SubsMat.MatrixInfo import blosum62
-    print(blosum62)
-except ModuleNotFoundError:
     from Bio.Align import substitution_matrices
     blosum62 = substitution_matrices.load("BLOSUM62")
     _tmp = {}
@@ -46,7 +43,8 @@ except ModuleNotFoundError:
         for ab2 in blosum62.alphabet:
             _tmp[(ab1, ab2)] = int(blosum62[(ab1, ab2)])
     blosum62 = _tmp
-    print(blosum62)
+except ModuleNotFoundError:
+    from Bio.SubsMat.MatrixInfo import blosum62
 
 AAstring = "ACDEFGHIKLMNPQRSTVWY"
 AAstringList = list(AAstring)

biopipen/scripts/tcr/ScRepCombiningExpression.R

@@ -7,7 +7,7 @@ srtobjfile <- {{in.srtobj | r}}
 outfile <- {{out.outfile | r}}
 cloneCall <- {{envs.cloneCall | r}}
 chain <- {{envs.chain | r}}
-group.by <- {{envs["group-by"] | r}}
+group_by <- {{envs.group_by | default: envs["group-by"] | default: None | r}}
 proportion <- {{envs.proportion | r}}
 filterNA <- {{envs.filterNA | r}}
 cloneSize <- {{envs.cloneSize | r}}
@@ -28,12 +28,13 @@ obj <- combineExpression(
     sc.data = srtobj,
     cloneCall = cloneCall,
     chain = chain,
-    group.by = group.by,
+    group.by = group_by,
     proportion = proportion,
     filterNA = filterNA,
     cloneSize = unlist(cloneSize),
     addLabel = addLabel
 )
+obj$TCR_Presence <- !is.na(obj$CTaa)
 
 log$info("Saving combined object ...")
 save_obj(obj, outfile)

biopipen/scripts/tcr/ScRepLoading.R

@@ -118,8 +118,13 @@ load_contig <- function(input, sample, fmt) {
     fmt <- dirfmt[[2]]
     if (is.null(dir)) { return(NULL) }
     x <- loadContigs(dir, format = fmt %||% "10X")
-    x[[1]]$sample <- NULL
-    x[[1]]
+    x <- x[[1]]
+    x$sample <- NULL
+    if (identical(fmt %||% "10X", "10X") && colnames(x)[1] == "X") {
+        x$X <- NULL
+    }
+
+    x
 }
 
 

biopipen/scripts/tcr/TCRClustering.R

@@ -130,11 +130,10 @@ output.clusters_df.to_csv(clustcr_dir + "/clusters.txt", sep="\t", index=False)
     clustcr_file
 }
 
-clean_clustcr_output = function(clustcr_outfile, clustcr_input) {
+clean_clustcr_output = function(clustcr_outfile) {
     clustcr_out = read.delim2(clustcr_outfile, header=TRUE, row.names = NULL)
     colnames(clustcr_out) = c("CDR3.aa", "TCR_Cluster")
-    in_cdr3 = read.delim2(clustcr_input, header=TRUE, row.names = NULL)
-    out = left_join(in_cdr3, distinct(clustcr_out), by=c("CDR3.aa")) %>%
+    out = left_join(cdr3aa_df, distinct(clustcr_out), by=c(cdr3seq4clustering = "CDR3.aa")) %>%
         mutate(
             TCR_Cluster = if_else(
                 is.na(TCR_Cluster),
@@ -170,7 +169,7 @@ run_clustcr = function() {
         quit(status=rc)
     }
     clustcr_outfile = file.path(clustcr_dir, "clusters.txt")
-    clean_clustcr_output(clustcr_outfile, clustcr_input)
+    clean_clustcr_output(clustcr_outfile)
 }
 
 prepare_giana = function() {
@@ -193,21 +192,8 @@ prepare_giana = function() {
 }
 
 prepare_input = function() {
-    # prepare input file for GIANA
-    cdr3 = c()
-    # cdr3col = if (!on_multi) "cdr3" else "CDR3.aa"
-    cdr3col = "CDR3.aa"
-    for (sample in names(seqdata)) {
-        sdata = seqdata[[sample]]
-        if (on_multi) {
-            sdata[[cdr3col]] = sub(";", "", sdata[[cdr3col]])
-        } else if ("chain" %in% colnames(sdata)) {
-            sdata = sdata %>% separate_rows(chain, cdr3col, sep = ";") %>%
-                filter(chain == "TRB")
-        }
-        cdr3 = union(cdr3, unique(sdata[[cdr3col]]))
-    }
-    cdr3 = unique(cdr3)
+    cdr3aa_df$cdr3seq4clustering <<- gsub("[^A-Z]", "", cdr3aa_df$CDR3.aa)  # Remove non-amino acid characters
+    cdr3 <- unique(cdr3aa_df$cdr3seq4clustering)
 
     # cdr3 = distinct(cdr3, aminoAcid, vMaxResolved)
 
@@ -220,15 +206,14 @@ prepare_input = function() {
     cdr3file
 }
 
-clean_giana_output = function(giana_outfile, giana_infile) {
+clean_giana_output = function(giana_outfile) {
     # generate an output file with columns:
     # CDR3.aa, TCR_Cluster, V.name, Sample
     # If sequence doesn't exist in the input file,
     # Then a unique cluster id is assigned to it.
     giana_out = read.delim2(giana_outfile, header=FALSE, comment.char = "#", row.names = NULL)[, 1:2, drop=FALSE]
     colnames(giana_out) = c("CDR3.aa", "TCR_Cluster")
-    in_cdr3 = read.delim2(giana_infile, header=TRUE, row.names = NULL)
-    out = left_join(in_cdr3, distinct(giana_out), by=c("CDR3.aa")) %>%
+    out = left_join(cdr3aa_df, distinct(giana_out), by=c(cdr3seq4clustering = "CDR3.aa")) %>%
         mutate(
             TCR_Cluster = if_else(
                 is.na(TCR_Cluster),
@@ -283,10 +268,11 @@ run_giana = function() {
         quit(status=rc)
     }
     giana_outfile = file.path(giana_outdir, "cdr3--RotationEncodingBL62.txt")
-    clean_giana_output(giana_outfile, giana_input)
+    clean_giana_output(giana_outfile)
 }
 
 attach_to_obj = function(obj, out) {
+    out <- as.data.frame(out)
     rownames(out) <- out$Barcode
     if (is_seurat) {
         # Attach results to Seurat object

biopipen/scripts/tcr/TESSA.R

@@ -39,9 +39,11 @@ log$info("Preparing TCR input file ...")
 # If immfile endswith .rds, then it is an immunarch object
 tcrdata <- sobj@meta.data %>%
     rownames_to_column("contig_id") %>%
+    select(contig_id, CTaa, CTgene, sample = Sample) %>%
     filter(!is.na(CTaa) & !is.na(CTgene)) %>%
-    separate(CTaa, into = c(NA, "cdr3"), sep = "_", remove = FALSE) %>%
-    separate(CTgene, into = c(NA, "vjgene"), sep = "_", remove = FALSE) %>%
+    separate(CTaa, into = c(NA, "cdr3"), sep = "_", remove = TRUE) %>%
+    filter(!is.na(cdr3) & cdr3 != "NA" & cdr3 != "nan") %>%
+    separate(CTgene, into = c(NA, "vjgene"), sep = "_", remove = TRUE) %>%
     separate(vjgene, into = c("v_gene", NA, "j_gene", NA), sep = "\\.", remove = TRUE) %>%
     mutate(v_gene = sub("-\\d+$", "", v_gene), j_gene = sub("-\\d+$", "", j_gene))
 

{biopipen-0.34.1.dist-info → biopipen-0.34.3.dist-info}/METADATA

@@ -1,6 +1,6 @@
 Metadata-Version: 2.3
 Name: biopipen
-Version: 0.34.1
+Version: 0.34.3
 Summary: Bioinformatics processes/pipelines that can be run from `pipen run`
 License: MIT
 Author: pwwang