biopipen 0.28.1__py3-none-any.whl → 0.29.0__py3-none-any.whl

biopipen/scripts/cnv/AneuploidyScore.R

@@ -127,13 +127,32 @@ getCAA <- function(segf, cytoarm, tcn_col,
   return(as(seg_cyto_chr, "GRangesList"))
 }
 
-segments = read.table(segfile, header=T, row.names=NULL, sep="\t", stringsAsFactors=F)
-seg = data.frame(
-    seqnames = segments[, chrom_col],
-    start = segments[, start_col],
-    end = segments[, end_col],
-    seg.mean = segments[, seg_col]
-)
+if (endsWith(segfile, ".vcf") || endsWith(segfile, ".vcf.gz")) {
+  library(VariantAnnotation)
+  vcf = readVcf(segfile)
+  seg = data.frame(
+      seqnames = as.character(seqnames(vcf)),
+      start = start(vcf),
+      end = vcf@info[[end_col]],
+      seg.mean = vcf@info[[seg_col]]
+  )
+} else if (endsWith(segfile, ".bed")) {
+  segments = read.table(segfile, header=F, row.names=NULL, sep="\t", stringsAsFactors=F)
+  seg = data.frame(
+      seqnames = segments[, 1],
+      start = segments[, 2],
+      end = segments[, 3],
+      seg.mean = segments[, 5]
+  )
+} else {
+  segments = read.table(segfile, header=T, row.names=NULL, sep="\t", stringsAsFactors=F)
+  seg = data.frame(
+      seqnames = segments[, chrom_col],
+      start = segments[, start_col],
+      end = segments[, end_col],
+      seg.mean = segments[, seg_col]
+  )
+}
 
 {% if envs.segmean_transform %}
 segmean_transform = {{envs.segmean_transform}}
@@ -168,6 +187,10 @@ if (is.character(cn_transform)) {
 }
 {% endif %}
 
+seg <- seg[
+  !is.na(seg$seg.mean) & !is.na(seg$TCN) & !is.infinite(seg$seg.mean) & !is.infinite(seg$TCN),,
+  drop=FALSE]
+
 write.table(seg, file.path(outdir, "seg.txt"), sep="\t", quote=F, row.names=F, col.names=T)
 
 wgd_ploidy = checkIfWGD(

biopipen/scripts/cnv/AneuploidyScoreSummary.R

@@ -52,8 +52,11 @@ if (!is.null(group_cols)) {
 
 if (!is.null(metafile)) {
     metadf = read.table(metafile, header=T, row.names=NULL, sep="\t", stringsAsFactors=F)
-    sample_col = colnames(metadf)[1]
-    colnames(metadf)[1] = "Sample"
+    if (!is.null(metadf$Sample)) {
+        metadf$Sample = as.character(metadf$Sample)
+    } else {
+        colnames(metadf)[1] = "Sample"
+    }
     metadf = metadf[metadf$Sample %in% sams, c("Sample", meta_cols), drop=FALSE]
     if (nrow(metadf) != length(sams)) {
         stop(paste("Not all samples in metafile:", paste(setdiff(sams, metadf$Sample), collapse=", ")))

biopipen/scripts/cnv/TMADScore.R

@@ -11,11 +11,27 @@ if (is.character(segmean_transform)) {
     segmean_transform = eval(parse(text=segmean_transform))
 } # otherwise NULL
 
-segments = read.table(segfile, header=T, row.names=NULL, sep="\t", stringsAsFactors=F)
-seg = data.frame(
-    chrom = segments[, chrom_col],
-    log2 = segments[, seg_col]
-)
+
+if (endsWith(segfile, ".vcf") || endsWith(segfile, ".vcf.gz")) {
+  library(VariantAnnotation)
+  segments = readVcf(segfile)
+  seg = data.frame(
+      chrom = as.character(seqnames(segments)),
+      log2 = segments@info[[seg_col]]
+  )
+} else if (endsWith(segfile, ".bed")) {
+  segments = read.table(segfile, header=F, row.names=NULL, sep="\t", stringsAsFactors=F)
+  seg = data.frame(
+      chrom = segments[, 1],
+      log2 = segments[, 5]
+  )
+} else {
+  segments = read.table(segfile, header=T, row.names=NULL, sep="\t", stringsAsFactors=F)
+  seg = data.frame(
+      chrom = segments[, chrom_col],
+      log2 = segments[, seg_col]
+  )
+}
 rm(segments)
 
 if (!is.null(excl_chroms) && length(excl_chroms) > 0) {

biopipen/scripts/cnv/TMADScoreSummary.R

@@ -49,8 +49,12 @@ if (!is.null(group_cols)) {
 data = data.frame(Sample = sams, tMAD = tmads)
 if (file.exists(metafile) && length(meta_cols) > 0) {
     metadf = read.table(metafile, header=T, row.names=NULL, sep="\t", stringsAsFactors=F)
-    sample_col = colnames(metadf)[1]
-    meta = metadf[, c(sample_col, meta_cols), drop=FALSE]
+    if (!is.null(metadf$Sample)) {
+        metadf$Sample = as.character(metadf$Sample)
+    } else {
+        colnames(metadf)[1] = "Sample"
+    }
+    meta = metadf[, c("Sample", meta_cols), drop=FALSE]
     colnames(meta) = c("Sample", meta_cols)
     data = data %>% left_join(meta, by="Sample")
 }

biopipen/scripts/cnvkit/CNVkitAccess.py

@@ -1,3 +1,4 @@
+from pathlib import Path
 from biopipen.utils.misc import run_command, dict_to_cli_args
 
 excfiles = {{in.excfiles | repr}}  # pyright: ignore
@@ -12,7 +13,7 @@ def main():
         "": [cnvkit, "access"],
         "s": min_gap_size,
         "o": outfile,
-        "_": reffile,
+        "_": Path(reffile).expanduser(),
     }
     if excfiles:
         other_args["exclude"] = excfiles

biopipen/scripts/cnvkit/CNVkitAutobin.py

@@ -1,3 +1,4 @@
+from pathlib import Path
 from biopipen.utils.misc import run_command, dict_to_cli_args
 
 bamfiles = {{in.bamfiles | repr}}  # pyright: ignore
@@ -20,7 +21,7 @@ short_names = {{envs.short_names | repr}}  # pyright: ignore
 def main():
 
     args = dict(
-        f=reffile,
+        f=Path(reffile).expanduser(),
         m=method,
         g=accfile,
         t=baitfile,
@@ -29,7 +30,7 @@ def main():
         target_min_size=target_min_size,
         antitarget_max_size=antitarget_max_size,
         antitarget_min_size=antitarget_min_size,
-        annotate=annotate,
+        annotate=Path(annotate).expanduser(),
         short_names=short_names,
         target_output_bed=target_file,
         antitarget_output_bed=antitarget_file,

biopipen/scripts/cnvkit/CNVkitBatch.py

@@ -42,7 +42,7 @@ def gen_access():
         exclude=access_excludes or False,
         s=access_min_gap_size or False,
         o=accessfile,
-        _=ref,
+        _=Path(ref).expanduser(),
     )
     args[""] = [cnvkit, "access"]
     run_command(dict_to_cli_args(args, dashify=True), fg=True)

biopipen/scripts/cnvkit/CNVkitCoverage.py

@@ -1,3 +1,4 @@
+from pathlib import Path
 from biopipen.utils.misc import run_command, dict_to_cli_args
 
 bamfile = {{in.bamfile | quote}}  # pyright: ignore
@@ -13,7 +14,7 @@ ncores = {{envs.ncores | repr}}  # pyright: ignore
 def main():
 
     args = dict(
-        f=reffile,
+        f=Path(reffile).expanduser(),
         c=count,
         q=min_mapq,
         p=ncores,

biopipen/scripts/cnvkit/CNVkitGuessBaits.py

@@ -60,7 +60,7 @@ params.update({
     "o": targetfile,
     "c": covfile,
     "p": ncores,
-    "f": ref,
+    "f": Path(ref).expanduser(),
     "s": samtools,
     "_": bamfiles,
 })

biopipen/scripts/cnvkit/CNVkitHeatmap.py

@@ -4,7 +4,7 @@ from diot import Diot
 
 from biopipen.utils.misc import run_command, dict_to_cli_args
 
-segfiles = {{in.segfiles | repr}}  # pyright: ignore
+segfiles = {{in.segfiles | repr}}  # pyright: ignore # noqa
 sample_sex = {{in.sample_sex | repr}}  # pyright: ignore
 outdir = {{out.outdir | repr}}  # pyright: ignore
 cnvkit = {{envs.cnvkit | quote}}  # pyright: ignore

biopipen/scripts/cnvkit/CNVkitReference.py

@@ -1,3 +1,4 @@
+from pathlib import Path
 from biopipen.utils.misc import run_command, dict_to_cli_args
 
 covfiles = {{in.covfiles | repr}}  # pyright: ignore
@@ -18,7 +19,7 @@ no_rmask = {{envs.no_rmask | repr}}  # pyright: ignore
 def main():
 
     args = dict(
-        f=reffile,
+        f=Path(reffile).expanduser(),
         o=outfile,
         c=cluster,
         min_cluster_size=min_cluster_size,

biopipen/scripts/gene/GeneNameConversion.R

@@ -0,0 +1,65 @@
+source("{{biopipen_dir}}/utils/misc.R")
+source("{{biopipen_dir}}/utils/gene.R")
+
+infile <- {{in.infile | quote}}
+outfile <- {{out.outfile | quote}}
+notfound <- {{envs.notfound | r}}
+genecol <- {{envs.genecol | r}}
+output <- {{envs.output | r}}
+dup <- {{envs.dup | r}}
+infmt <- {{envs.infmt | r}}
+outfmt <- {{envs.outfmt | r}}
+species <- {{envs.species | r}}
+
+if (is.na(notfound)) {
+    notfound = "na"
+}
+
+df <- read.table(infile, header=TRUE, sep="\t", check.names=FALSE)
+
+if (genecol == 0) {
+    log_warn("envs.genecol should be 1-based, but 0 was given. Using 1 instead.")
+    genecol <- 1
+}
+
+if (is.numeric(genecol)) { genecol <- colnames(df)[genecol] }
+if (dup == "combine") { dup <- ";" }
+
+genes <- df[[genecol]]
+converted <- gene_name_conversion(
+    genes=genes,
+    species=species,
+    infmt=infmt,
+    outfmt=outfmt,
+    notfound=notfound,
+    dup=dup
+)
+#    <genecol> <outfmt>
+# 1  1255_g_at   GUCA1A
+# 2    1316_at     THRA
+# 3    1320_at   PTPN21
+# 4    1294_at  MIR5193
+
+# order the converted dataframe by the original gene column
+converted <- converted[order(match(converted$query, genes)), , drop=FALSE]
+outcol <- outfmt
+
+if (notfound == "skip" || notfound == "ignore") {
+    df <- df[df[[genecol]] %in% converted$query, , drop=FALSE]
+}
+
+if (output == "append") {
+    if (outfmt %in% colnames(df)) {
+        log_warn("The output column name already exists in the input dataframe. Appending with a suffix `_1`.")
+        outcol <- paste(outfmt, "_1", sep="")
+    }
+    df[[outcol]] <- converted[[outfmt]]
+} else if (output == "replace") {
+    df[[genecol]] <- converted[[outfmt]]
+} else if (output == "with-query") {
+    df <- converted
+} else {
+    df <- converted[, outfmt, drop=FALSE]
+}
+
+write.table(df, file=outfile, sep="\t", quote=FALSE, row.names=FALSE)

biopipen/scripts/gene/GenePromoters.R

@@ -0,0 +1,61 @@
+library(rlang)
+library(rtracklayer)
+
+infile <- {{in.infile | r}}
+outfile <- {{out.outfile | r}}
+up <- {{envs.up | r}}
+down <- {{envs.down | r}}
+notfound <- {{envs.notfound | r}}
+refgene <- {{envs.refgene | r}}
+header <- {{envs.header | r}}
+genecol <- {{envs.genecol | r}}
+match_id <- {{envs.match_id | r}}
+sort_ <- {{envs.sort | r}}
+chrsize <- {{envs.chrsize | r}}
+
+down <- down %||% up
+
+refgenes <- readGFF(refgene)
+refcol <- ifelse(match_id, "gene_id", "gene_name")
+
+if (infile == "/dev/null") {
+    genes <- unique(refgenes[[refcol]])
+} else {
+    data <- read.table(infile, header=header, sep="\t", stringsAsFactors=FALSE, check.names=FALSE)
+    genes <- data[[genecol]]
+    rm(data)
+}
+
+notfound_genes <- setdiff(genes, refgenes[[refcol]])
+if (notfound == "error" && length(notfound_genes) > 0) {
+    stop(paste(
+        "The following genes were not found in the reference annotation:",
+        paste(notfound_genes, collapse=", ")
+    ))
+} else if (notfound == 'skip') {
+    genes <- genes[!genes %in% notfound_genes]
+}
+
+# Select the genes that are in the reference annotation and keep the order
+# of the records in genes
+refgenes <- refgenes[match(genes, refgenes[[refcol]]), , drop = FALSE]
+refgenes <- unique(makeGRangesFromDataFrame(refgenes, keep.extra.columns=TRUE))
+
+proms <- promoters(refgenes, up=up, down=down)
+# Scores must be non-NA numeric values
+elementMetadata(proms)$name <- elementMetadata(proms)[[refcol]]
+score(proms) <- 0
+start(proms) <- pmax(1, start(proms))
+
+if (sort_) {
+    chrom_sizes <- read.table(chrsize, header=FALSE, stringsAsFactors=FALSE, sep="\t")
+    common_chroms <- intersect(chrom_sizes$V1, seqlevels(proms))
+    if (length(common_chroms) == 0) {
+        stop("No common chromosomes found between the promoters and the chromosome sizes. Do you use the correct chromosome sizes file?")
+    }
+    proms <- keepSeqlevels(proms, common_chroms, pruning.mode="coarse")
+    seqlevels(proms) <- common_chroms
+    proms <- sort(proms, ignore.strand = TRUE)
+}
+
+export.bed(proms, outfile)

biopipen/scripts/misc/Shell.sh

@@ -0,0 +1,15 @@
+# shellcheck disable=all
+export infile={{in.infile | quote}}
+export outfile={{out.outfile | quote}}
+is_outdir={{envs.outdir | int}}
+cmd_given={{envs.cmd | bool | int}}
+{% set _ = out.outfile | dirname | joinpath: "cmd.sh" | as_path | attr: 'write_text' | call: envs.cmd %}
+cmd="{{proc.lang}} {{out.outfile | dirname | joinpath: 'cmd.sh'}}"
+if [[ "$cmd_given" -eq 0 ]]; then
+    echo "No command given." 1>&2
+    exit 1
+fi
+if [[ $is_outdir -eq 1 ]]; then
+    mkdir -p "$outfile"
+fi
+eval "$cmd"

biopipen/scripts/plot/Manhattan.R

@@ -0,0 +1,140 @@
+source("{{biopipen_dir}}/utils/misc.R")
+library(rlang)
+library(ggmanh)
+
+infile <- {{in.infile | r}}
+outfile <- {{out.outfile | r}}
+chrom_col <- {{envs.chrom_col | r}}
+pos_col <- {{envs.pos_col | r}}
+pval_col <- {{envs.pval_col | r}}
+label_col <- {{envs.label_col | r}}
+devpars <- {{envs.devpars | r}}
+title <- {{envs.title | r}}
+ylabel <- {{envs.ylabel | r}}
+rescale <- {{envs.rescale | r}}
+rescale_ratio_threshold <- {{envs.rescale_ratio_threshold | r}}
+signif <- {{envs.signif | r}}
+hicolors <- {{envs.hicolors | r}}
+thin_n <- {{envs.thin_n | r}}
+thin_bins <- {{envs.thin_bins | r}}
+zoom <- {{envs.zoom | r}}
+zoom_devpars <- {{envs.zoom_devpars | r}}
+chroms <- {{envs.chroms | r}}
+args <- {{envs.args | r: todot="-"}}
+
+data <- read.table(infile, header=TRUE, sep="\t", stringsAsFactors=FALSE, check.names = FALSE)
+
+# normalize columns
+cnames <- colnames(data)
+if (is.numeric(chrom_col)) { chrom_col <- cnames[chrom_col] }
+if (is.numeric(pos_col)) { pos_col <- cnames[pos_col] }
+if (is.numeric(pval_col)) { pval_col <- cnames[pval_col] }
+if (is.numeric(label_col)) { label_col <- cnames[label_col] }
+
+# normalize chroms
+norm_chroms <- function(chrs) {
+    chrs <- as.character(chrs)
+    if (length(chrs) == 1 && grepl(",", chrs)) {
+        chrs <- trimws(unlist(strsplit(chrs, ",")))
+    }
+    if (length(chrs) > 1) {
+        return(unique(unlist(sapply(chrs, function(chr) norm_chroms(chr)))))
+    }
+    if (!grepl("-", chrs)) { return(chrs) }
+
+    # expand chr1-22 -> chr1, chr2, ..., chr22
+    # chr1-22 -> 'chr1', '22'
+    chrs <- unlist(strsplit(chrs, "-"))
+    if (length(chrs) != 2) {
+        stop(paste0("Invalid chroms: ", chrs))
+    }
+    # detect prefix
+    prefix1 <- gsub("[0-9]", "", chrs[1])
+    prefix2 <- gsub("[0-9]", "", chrs[2])
+    if (nchar(prefix2) > 0 && prefix1 != prefix2) {
+        stop(paste0("Invalid chroms: ", chrs, " (prefix mismatch)"))
+    }
+    chr_a <- as.integer(substring(chrs[1], nchar(prefix1) + 1))
+    chr_b <- as.integer(substring(chrs[2], nchar(prefix2) + 1))
+    chr_min <- min(chr_a, chr_b)
+    chr_max <- max(chr_a, chr_b)
+    return(paste0(prefix1, chr_min:chr_max))
+}
+
+log_info("Preparing data for plotting ...")
+if (length(chroms) == 1 && chroms == "auto") {
+    chroms <- unique(data[[chrom_col]])
+} else {
+    chroms <- norm_chroms(chroms)
+}
+
+# prepare data
+mp_prep_args = list()
+if (length(signif) == 1 && is.character(signif)) {
+    signif <- as.numeric(trimws(unlist(strsplit(signif, ","))))
+}
+siglevel <- min(signif)
+if (!is.null(label_col)) {
+    data$.label <- ifelse(data[[pval_col]] < siglevel, data[[label_col]], "")
+}
+if (!is.null(hicolors)) {
+    sig_str <- "Significant"
+    nsig_str <- "Not significant"
+    data$.highlight <- ifelse(data[[pval_col]] < siglevel, sig_str, nsig_str)
+    if (length(hicolors) == 1) { hicolors <- c(hicolors, "grey") }
+    names(hicolors) <- c(sig_str, nsig_str)
+    mp_prep_args$highlight.colname <- ".highlight"
+    mp_prep_args$highlight.col <- hicolors
+}
+mp_prep_args$x <- data
+mp_prep_args$chr.colname <- chrom_col
+mp_prep_args$pos.colname <- pos_col
+mp_prep_args$pval.colname <- pval_col
+mp_prep_args$chr.order <- chroms
+if (!is.null(thin_n) && thin_n > 0) {
+    mp_prep_args$thin.n <- thin_n
+    mp_prep_args$thin.bins <- thin_bins
+}
+
+mpdata <- do_call(manhattan_data_preprocess, mp_prep_args)
+
+# plot
+log_info("Plotting Manhattan plot ...")
+args$x <- mpdata
+args$signif <- signif
+args$plot.title <- title
+args$rescale <- rescale
+args$rescale.ratio.threshold <- rescale_ratio_threshold
+if (!is.null(hicolors)) { args$color.by.highlight <- TRUE }
+if (!is.null(label_col)) { args$label.colname <- ".label" }
+g <- do_call(manhattan_plot, args)
+
+png(outfile, width=devpars$width, height=devpars$height, res=devpars$res)
+print(g)
+dev.off()
+
+# zoom into chromosomes
+if (!is.null(zoom)) {
+    log_info("Zooming into chromosomes ...")
+    zoom <- norm_chroms(zoom)
+    for (z in zoom) {
+        log_info("- {z}")
+        args_z <- args
+        args_z$chromosome <- z
+        args_z$plot.title <- paste0(title, " (", z, ")")
+        args_z$x.label <- "Position"
+        g_z <- do_call(manhattan_plot, args_z)
+        outfile_z <- gsub("\\.png$", paste0("-", z, ".png"), outfile)
+        zm_devpars <- zoom_devpars
+        zm_devpars$res <- zm_devpars$res %||% devpars$res
+        zm_devpars$height <- zm_devpars$height %||% devpars$height
+        png(
+            outfile_z,
+            width=zm_devpars$width,
+            height=zm_devpars$height,
+            res=zm_devpars$res
+        )
+        print(g_z)
+        dev.off()
+    }
+}

biopipen/scripts/plot/QQPlot.R

@@ -0,0 +1,62 @@
+source("{{biopipen_dir}}/utils/misc.R")
+
+library(ggplot2)
+library(ggprism)
+library(qqplotr)
+
+theme_set(theme_prism())
+
+infile <- {{in.infile | r}}
+outfile <- {{out.outfile | r}}
+val_col <- {{envs.val_col | r}}
+devpars <- {{envs.devpars | r}}
+title <- {{envs.title | r}}
+xlabel <- {{envs.xlabel | r}}
+ylabel <- {{envs.ylabel | r}}
+kind <- {{envs.kind | r}}
+trans <- {{envs.trans | r}}
+band_args <- {{envs.band | r}}
+line_args <- {{envs.line | r}}
+point_args <- {{envs.point | r}}
+ggs <- {{envs.ggs | r}}
+
+indata <- read.table(infile, header=TRUE, sep="\t", stringsAsFactors=FALSE, check.names = FALSE)
+if (is.numeric(val_col)) { val_col <- colnames(indata)[val_col] }
+
+band_fun <- ifelse(kind == "pp", stat_pp_band, stat_qq_band)
+line_fun <- ifelse(kind == "pp", stat_pp_line, stat_qq_line)
+point_fun <- ifelse(kind == "pp", stat_pp_point, stat_qq_point)
+
+title <- title %||% waiver()
+xlabel <- xlabel %||% waiver()
+ylabel <- ylabel %||% waiver()
+
+if (!is.null(trans)) {
+    trans <- trimws(trans)
+    if (trans == "-log10") {
+        trans <- function(x) -log10(x)
+    } else {
+        trans <- eval(parse(text = trans))
+    }
+
+    indata$.trans_val <- trans(indata[[val_col]])
+    val_col <- ".trans_val"
+}
+
+indata <- indata[!is.na(indata[[val_col]]), , drop=FALSE]
+
+p <- ggplot(data = indata, mapping = aes(sample = !!sym(val_col))) +
+    do_call(band_fun, band_args) +
+    do_call(line_fun, line_args) +
+    do_call(point_fun, point_args) +
+    labs(title = title, x = xlabel, y = ylabel)
+
+if (!is.null(ggs)) {
+    for (gg in ggs) {
+        p <- p + eval(parse(text = gg))
+    }
+}
+
+png(outfile, width=devpars$width, height=devpars$height, res=devpars$res)
+print(p)
+dev.off()