4.2 BAMDiagnostics

Estimating approximate depth, read length frequencies and mapping quality frequencies

BAMDiagnostics provides a set of read statistics for the input BAM file while taking into account all standard input filters. The output are written to .txt files that summarize the following information:

• Total number of reads
• Number of reads that passed filters
• Number of duplicate reads
• Average read length
• Maximum read length
• Number of proper pairs
• Average fragment length (only known for paired-end data)
• Total number of soft-clipped positions
• Average soft-clipped length
• Average aligned length
• Mean sequencing depth across the whole genome
• Average mapping quality

It also provides histograms which display the distributions of fragment lengths, mapping qualities, read lengths, soft-clipped lengths and aligned lengths. All of this data is written for all read groups combined, as well as for each read group separately.

4.2.1 Input

Required inputs :

--bam Input_bam_file.bam

Input BAM file.

Optional inputs :

• none

Specific Parameters :

--diagnosticsPerChromosome	To output data per chromosome into a *_diagnostics.txt diagnostics file. Default = Only per-read group summary statistics is provided (per chromosome summary statistics is provided).
--splitMergeInput	To create input file for splitMerge. Default = Will not create input file for splitMerge.
--printReferenceLength	To print reference lengths of chromosomes to file. Default = Will not print reference lengths of chromosomes to file.

• See Filter parameters to apply specific filters for bases, reads and parsing window setting.

Engine parameters that are common to all tasks can be found here.

4.2.2 Output

*_filterSummary.txt	Filter summary for all read groups combined and individual read groups.
*_fragmentLengthHistogram.txt	Counts for all fragment length for all read groups combined and individual read groups.
*_mappingQualityHistogram.txt	Mapping quality counts for all read groups combined and individual read groups.
*_readLengthHistogram.txt	Read length counts for all read groups combined and individual read groups.
*_softClippedLengthHistogram.txt	Length of soft-clipped bases as counts for all read groups combined and individual read groups.
*_alignedLengthHistogram.txt	Aligned length counts for all read groups combined and individual read groups
*_diagnostics.txt	File containing per-read group summary statistics. Also contains per chromosome summary statistics is provided when --diagnosticsPerChromosome parameter is used. This file can be used as input file for the splitMerge task.

4.2.3 Usage Example

#! /bin/bash

# `--fixedSeed = N` is needed to have reproducable results in regression test

. $(dirname $0)/find_atlas
. $(dirname $0)/simulate --type HW --F 0.1 --fixedSeed 0 \
    --sampleSize 17 --chrLength 11111 --fracPoly 1.0 \
    --alpha 2.0 --beta 2.0 --seqType single --seqCycles 101

for i in {1..17}; do
    samtools view simulate_ind"$i".bam | head -250 | tail -10 | cut -f1 \
            > blacklist_"$i".txt
    u=$(echo "$i*5" | bc)

    out="simple$i"
    $atlas --task BAMDiagnostics --perChromosome --bam simulate_ind$i.bam \
           --fixedSeed $i --out $out --logFile $out.out 2> $out.eout

    out="complex$i"
    $atlas --task BAMDiagnostics --identifyDuplicates --bam simulate_ind$i.bam \
           --filterSoftClips --filterMQ 0,$u --blacklist blacklist_$i.txt \
           --filterReadLength 0,$u --filterFragmentLength 0,$u \
           --fixedSeed 1$i --out $out --logFile $out.out 2> $out.eout
done