From a03e1e736a94f71c043df6ef205ea4bb1364570c Mon Sep 17 00:00:00 2001
From: =?UTF-8?q?Miguel=20=C3=81lvarez=20Herrera?= <miguel.alvarez@uv.es>
Date: Mon, 24 Nov 2025 13:38:08 +0100
Subject: [PATCH] fix: force reading VCF fields as character

Co-authored-by: Lluna Bellot <llunabellotpastor@gmail.com>
---
 workflow/scripts/format_vcf_fields_longer.R | 10 ++++++++--
 1 file changed, 8 insertions(+), 2 deletions(-)

diff --git a/workflow/scripts/format_vcf_fields_longer.R b/workflow/scripts/format_vcf_fields_longer.R
index 26c5512..a1584f3 100644
--- a/workflow/scripts/format_vcf_fields_longer.R
+++ b/workflow/scripts/format_vcf_fields_longer.R
@@ -27,13 +27,19 @@ filter.exclude <- lapply(snakemake@params$filter_exclude, empty.to.na)
 
 # Process input table
 log_info("Applying filters and writing results")
-read_tsv(snakemake@input$tsv) %>%
+read_tsv(
+  snakemake@input$tsv,
+  col_types = cols(
+    POS = col_integer(),
+    .default = col_character()
+  )
+) %>%
 
   # Separate <sep>-delimited "...[*]..." columns (e.g. ANN[*].EFFECT)
   separate_rows(
     contains("[*]"),
     sep = snakemake@params$sep,
-    convert = TRUE
+    convert = FALSE
   ) %>%
 
   # Rename "...[*]..." columns using the provided lookup via Snakemake config