Shi Xing post-process from demultiplex to rearr

#!/bin/bash

shopt -s expand_aliases

alias ~=”:«’~bash’”

:«’~~~bash’

Usage

sxCutR2AdapterFilterCumulate.md demultiplexFile minToMapShear >toMapFile

Introduction

The output of demultiplex.md needs further post-process before feed to rearr. For Shi Xing’s data, this is done by this in-house script. The output format fits the input format of rearr.

query<tab>count<tab>refId<newline>

The post-process consists of three steps.

1. Remove adapter from 3’ of R2.
2. Remove primer, barcode and a 3bp gap from 5’ of R2.
3. Filter out if the remain of R2 is shorter than minToMapShear.
4. Accumulate the adjacent duplicates by sxCumulateToMapCutAdaptSpliter.awk.

Source

cutadaptPlain()
{
    # Usage: cutadaptPlain <plainseq 3'adapter
    # cutadapt does not accept plainseq. This function transform plainseq to fasta before feed to cutadapt, and then transform the fasta output back to plainseq
    # Input: plainseq
    # Output: 3' trimmed plainseq
    sed '=' | sed '1~2s/^/>s/' | cutadapt -a $1 - 2> /dev/null | sed '1~2d'
}

rmDupFile=$1
minToMapShear=$2
cut -f1 $rmDupFile | cutadaptPlain GCACCGACTCGGTGCCACTTTTTCAAGTTGATAACGGACTAGCCTTATTTTAACTTGCTATTTCTAGCTCTAAAAC | paste - <(cut -f3-4,8 $rmDupFile) | gawk -F "\t" -v OFS="\t" -v minToMapShear=$minToMapShear '
{
    if ($4 + minToMapShear <= length($1)) {
        print substr($1, $4 + 1), $2, $3
    }
}' | gawk -f sxCumulateToMapCutAdaptSpliter.awk

alias ~~~=":" # This suppresses a warning and is not part of source.