Why MAQ consensus seq better than SAMtools consensus ??

[av_d]

Hi,
I was trying to assemble whole genome illumina 36 bp reads using MAQ and generate consensus fastq seq. I've also used SAMtool to call SNP/indel and generate consensus seq from .map output of MAQ. But the quality of the consensus seq from MAQ is far better than that from SAMtool. What is wrong with SAMtools?

Following SAMtool cmd used:

maq2sam-long final.map > final.sam
samtools view -bSt ref.fa.fai -o final.bam final.sam
samtools sort final.bam final.sorted
samtools index final.sorted.bam
samtools pileup -vcf ref.fa final.sorted.bam > final.sorted.bam.cns
samtools.pl pileup2fq -D 100 final.sorted.bam.cns > final.sorted.bam.fq


First 20 lines from MAQ consensus fastq:
@I
nnnnnnnnnnnnnnnnnnnnnnnnnnnnnnnnnnnnnnnnnnnnnnnnnnnnnnnnnnnn
nnnnnnnnnnnnnnnnnnnnnnnnnnnnnnnnnnnnnnnnnnnnnnnnnnnnnnnnnnnn
nnnnnnnnnnnnnnnnnnnnnnnnnnnnnnnnnnnnnnnnnnnnnnnnnnnnnnnnnnnn
nnnnnnnnnnnnnnnnnnnnnnnnnnnnnnnnnnnnnnnnnnnnnnnnnnnnnnnnnnnn
nnnnnnnnnnnnnnnnnnnnnnnnnnnnnnnnnnnnnnnnnnnnnnnnnnnnnnnnnnnn
nnnnnnnnnnnnnnnnnnnnnnnnnnnnnnnnnnnnnnnnnnnnnnnnnnnnnnnnnnnn
nnnnnnnnnnnnnnnnnnnnnnnnnnnnnnnnnnnnnnnnnnnnnnnagcCTAAGCCTAA
GCCTAAGCCTAAAAAATTGAGATAAGAAAACATTTTACTTtttcannnnnnnnnnnnnnn
nnnnnnnnnnnnnnnnnnnnnnnnnnnnnnnnnnnnnnnnnnnnnnnnnnnnnnnnnnnn
nnnnnnnnnngccaacctatatgctcctgtgtttaggcctaATACTAAGCCTAAGCctaa
gcctaatactaagcctannnnnnnnnnnnnnnnnnnnnnnnnnnnnnnnnnnnnnnnnnn
nnnnnnnnnnnnnnnnnnnnnnnnnnnnnnnnnnnnnnnnnngcctaagactaagcctaa
gcctaatactaagcctaagnnnnnnnnnnnnnnnnnnnnnnnnnnnnnnnnnnnnnnnnn
nnnnnnnnnnnnnnnnnnnnnnnnnnnnnnnactaagcctaagcctaagactaAgccTaa
gCCtaaAagaATATGGTAGCTACAGAAACggtagtacactcttctgaaagnnnnnnnnnn
ttgcaatttttaTAGCTAGGGCACTTTTTGTCTGCCCAWATATAGGCAACCAAaaataat
tgccaagtttttaatgatttgttgcatattgannnnnnnnnnnnnnnnnnnnnnnnnnnn
nnntatcgtagctacagaAACGGTTGtgcActcATCTGAAAGTTTGTTTTTCTTgttttc
ttgcACTttgtgcagaATTCTTGATTCTTGATTCTTGCAGAAATTTGCAAGAaaattcgc

First 20 lines from SAMtools consensus fastq:

@I
nnnnnnnnnnnnnnnnnnnnnnnnnnnnnnnnnnnnnnnnnnnnnnnnnnnnnnnnnnnn
nnnnnnnnnnnnnnnnnnnnnnnnnnnnnnnnnnnnnnnnnnnnnnnnnnnnnnnnnnnn
nnnnnnnnnnnnnnnnnnnnnnnnnnnnnnnnnnnnnnnnnnnnnnnnnnnnnnnnnnnn
nnnnnnnnnnnnnnnnnnnnnnnnnnnnnnnnnnnnnnnnnnnnnnnnnnnnnnnnnnnn
nnnnnnnnnnnnnnnnnnnnnnnnnnnnnnnnnnnnnnnnnnnnnnnnnnnnnnnnnnnn
nnnnnnnnnnnnnnnnnnnnnnnnnnnnnnnnnnnnnnnnnnnnnnnnnnnnnnnnnnnn
nnnnnnnnnnnnnnnnnnnnnnnnnnnnnnnnnnnnnnnnnnnnnnnnnnnnnnnnnnnn
nnnnnnnnnnnnnnnnnnnnnnnnnnnnnnnnnnnnnnnnnnnnnnnnnnnnnnnnnnnn
nnnnnnnnnnnnnnnnnnnnnnnnnnnnnnnnnnnnnnnnnnnnnnnnnnnnnnnnnnnn
nnnnnnnnnnnnnnnnnnnnnnnnnnnnnnnnnnnnnnnnnnnnnnnnnnnnnnnnnnnn
nnnnnnnnnnnnnnnnnnnnnnnnnnnnnnnnnnnnnnnnnnnnnnnnnnnnnnnnnnnn
nnnnnnnnnnnnnnnnnnnnnnnnnnnnnnnnnnnnnnnnnnnnnnnnnnnnnnnnnnnn
nnnnnnnnnnnnnnnnnnnnnnnnnnnnnnnnnnnnnnnnnnnnnnnnnnnnnnnnnnnn
nnnnnnnnnnnnnnnnnnnnnnnnnnnnnnnnnnnnnnnnnnnnnnnnnnnnnnnnnnnn
nnnnnnnnnnnnnnnnnnnnnnnnnnnnnnnnnnnnnnnnnnnnnnnnnnnnnnnnnnnn
nnnnnnnnnnnnnnnnnnnnnnnnnnnnnnnnnnnnnnWnnnnnnnnnnnnnnnnnnnnn
nnnnnnnnnnnnnnnnnnnnnnnnnnnnnnnnnnnnnnnnnnnnnnnnnnnnnnnnnnnn
nnnnnnnnnnnnnnnnnnnnnnnnnnnnnnnnnnnnnnnnnnnnnnnnnnnnnnnnnnnn
nnnnnnnnnnnnnnnnnnnnnnnnnnnnnnnnnnnnnnnnnnnnnnnnnnnnnnnnnnnn

[av_d]

I got my answer.. "-v" should not be present with the "pileup" cmd.

[mixter]

Hi,

I have run into the same problem, however, your solution of removing the -v option in the samtools pileup call did not work for me. I still get sequences consisting of long stretches of "nnnnnnnnnnnnnnn" in the FASTQ files, resulting in conversion failing and empty FASTA files being generated out of them.

In my case, I am trying to generate consensus FASTA sequence from the mapped BAM files from the 1000 Genomes Project.

name=HG00096
file=HG00096.mapped.ILLUMINA.bwa.GBR.low_coverage.20101123
samtools view -bSt /opt/genomes/hg18-chromosomes.fa.fai -o ${file}.bam ${file}.sam
samtools sort ${file}.bam ${name}.final.sorted
samtools index ${name}.final.sorted.bam
samtools pileup -cf /gcm/opt/genomes/hg18-chromosomes.fa $file > ${name}.consensus.pileup
/gcm/opt/samtools/samtools-0.1.7a/misc/samtools.pl pileup2fq -D 100 ${name}.consensus.pileup > ${name}.fq
/gcm/opt/maq/maq-0.7.1/bin/fq_all2std.pl fq2fa ${name}.fastq > ${name}.fasta

Alternatively I tried the FASTX toolkit with no luck:
fastq_to_fasta -i ${name}.fq -o ${name}.fa


Hope someone can help

[av_d]

check your reference file first, because human ref genome already have long stretches of "nnnnnnnnnnnnnnn"

[gauravdube]

Hi av_d,

My fastq file consists of lot non-ATGC characters (you are too are getting in your file, you see that 'W' in your fastq ?). What are these characters and how to handle these?

Commands used:
bwa index ref.fa
bwa aln -t 9 cocsa_ref.fa D2_R2.fastq -f D2_R2.sai && bwa aln -t 9 cocsa_ref.fa D2_R1.fastq -f D2_R1.sai
bwa sampe ref.fa D2_R1.sai D2_R2.sai D2_R1.fq D2_R2.fq > D2-aln-pe2.sam
samtools faidx cocsa_ref.fa
samtools view -bt ref.fa.fai D2-aln-pe2.sam > D2-aln-pe2.bam
samtools sort D2-aln-pe2.bam D2-aln-pe2.bam.srt
samtools index D2-aln-pe2.bam.srt.bam
samtools mpileup -uf ref.fa D2-aln-pe2.bam.srt.bam | bcftools view -cg - | vcfutils.pl vcf2fq > CONSENSUS.fq


CONSENSUS.fq file looks like:
@scaffold_1
nnngtttggtggtagtattggtatttcaaacacgctaggtgtttgttggttttgagtagg
tgtagctggagtagactctatctccatttctctatcagtttgggcctctggccctaggct
ctcctgtctgttttcttgagtatttactacaatagtatcactgtctggcggcattttatt
actaagctcttttcttagtaagcaactagatggtctgtgtgtttttgttttcgtgagtga
gacgtgttcagattagctactttaccagcttctagctctatagcgcgtgggctgcacgag
ttggcactagttgtaatcgatttcttgggatggatttgtatataattcgctaaaattaca
cctattctgaaaaactcgnnnnnnnnnnnnnnnnnnnnnnnnnnnnnnnnnnnnnnnnnn
nnnnnnnnnnnnnnnnnnnnnnnnnnnnnnnnnnnnnnnnnnnnnnnnnnnnnnnnnnnn
nnnnnnnnTAATGTTACAAGTAAYAAGAAGGATYCTYTCCTTRACAAATRACGAGATGGC

Please also convey, how to handle the small-case characters and 'N's ?

Thanks in advance.

[profbiot]

The lowercase letters have poor mapping quality. The non-ATGC letters are IUPAC codes for positions with more than one base observed.

http://www.bioinformatics.org/sms2/iupac.html