Add rule to compile extracted VCF fields

Author: ahmig

workflow/rules/vaf.smk

@@ -181,6 +181,19 @@ rule format_vcf_fields_longer:
         "../scripts/format_vcf_fields_longer.R"
 
 
+rule compile_vcf_fields_longer:
+    threads: 1
+    conda: "../envs/renv.yaml"
+    input:
+        tsv = OUTDIR/"vaf"/"{sample}.vcf_fields.longer.tsv"
+    output:
+        tsv = OUTDIR/f"{OUTPUT_NAME}.vcf_fields.longer.tsv"
+    log:
+        LOGDIR / "compile_vcf_fields_longer" / "log.txt"
+    script:
+        "../scripts/compile_vcf_fields_longer.R"
+
+
 rule vcf_to_tsv:
     threads: 1
     conda: "../envs/renv.yaml"

workflow/scripts/compile_vcf_fields_longer.R

@@ -0,0 +1,23 @@
+#!/usr/bin/env Rscript
+
+# Write stdout and stderr to log file
+log <- file(snakemake@log[[1]], open = "wt")
+sink(log, type = "message")
+sink(log, type = "output")
+
+library(tidyverse)
+library(logger)
+log_threshold(INFO)
+
+log_info("Reading input partial tables with VCF extracted fields and adding sample column")
+tables <- lapply(
+    snakemake@input,
+    function(path) {
+        read_tsv(path) %>%
+            mutate(SAMPLE = sub('\\.vcf_fields.longer.tsv$', '', basename(path)))
+    }
+)
+
+log_info("Binding and writing table")
+bind_rows(tables) %>%
+    write_tsv(snakemake@output$tsv)