VCF format issues with --write-vcf, - FORMAT field inconsistencies about hap.py

[YuanTian1991]

My R code to parse the output:

library("data.table")
library("tidyverse")
library("reshape2")

message("Reading VCF...")
happy_vcf <- fread(happy_vcf_path, sep='\t', header = TRUE, skip = '#CHROM') %>%
    .[,INFO:=NULL]
  
  format_types <- unique(happy_vcf$FORMAT)
  
  truth <- list()
  query <- list()
  for(i in format_types) {
    message("Parsing ", i, "...")
    index <- which(happy_vcf$FORMAT == i)
    col_names <- str_split(i, ":")[[1]]
    
    truth[[i]] <- colsplit(happy_vcf[index, TRUTH], ":", names=col_names) %>% 
      setDT %>% .[,index:=index]
    
    query[[i]] <- colsplit(happy_vcf[index, QUERY], ":", names=col_names) %>% 
      setDT %>% .[,index:=index]
  }
  
  truth <- rbindlist(truth, fill = TRUE) %>% .[order(index), ]
  query <- rbindlist(query, fill = TRUE) %>% .[order(index), ]