fix bam merge

filter_bams.sh

 fd --extension bam --change-older-than "2020-02-26 00:00:00" | head -n 100

fix_merge.sh

#!/usr/bin/bash
# Author: Daniel E. Cook
# This script will fix merged IDs'
#SBATCH --job-name unmerge_bams
#SBATCH --part=cpu
#SBATCH --time=2:00:00
#SBATCH --cpus-per-task=6
#SBATCH --mem=4G
#SBATCH -o /camp/project/proj-tracerx-lung/tracerx/_PIPELINE/TRACERx-assets/logs/%j.out
#SBATCH -e /camp/project/proj-tracerx-lung/tracerx/_PIPELINE/TRACERx-assets/logs/%j.err
set -e

CORES=2

# Fix issue associated with bams that were merged without using the -p and -c flags
function fix_merged_bam
{
    # Output clean header
    {
        samtools view --threads ${CORES} -H ${1} | \
        grep -E -v 'ID:(bwa|MarkDuplicates|GATK IndelRealigner)(\.[0-9])?\-[A-Z0-9]+' | \
        grep -E -v 'ID:.*--[A-Za-z0-9]{12}-[0-9]{1}-[A-Z0-9]+';
        samtools view --threads ${CORES} ${1} | \
        sed -r 's/(RG.*--[A-Za-z0-9]{12}-[0-9]{1})-[A-Z0-9]+/\1/' | \
        sed -r 's/(MarkDuplicates)-[A-Z0-9]+/\1/'
    } | samtools view --threads ${CORES} -hb
}

# Fix bam
function fix_bam {
    temp_name=${1/.bam/.fix_merge.bam}
    >&2 echo "fixing ${1}; Writing to ${temp_name}"
    fix_merged_bam ${1} > ${temp_name}
    >&2 echo "fixing ${1}; Performing a quickcheck"
    samtools quickcheck ${temp_name}
    >&2 echo "Indexing ${temp_name}"
    samtools index --@ 6 ${1/.bam/.fix_merge.bam}
    >&2 echo "Moving: ${temp_name}"
    #mv ${temp_name} ${1}
    #mv ${temp_name}.bai ${1}.bai
    >&2 echo "Done: ${1} Replaced"
}

fix_bam ${1}