Go Back   SEQanswers > Bioinformatics > Bioinformatics

Similar Threads
Thread Thread Starter Forum Replies Last Post
Odd behavior using samtools mpileup, bcftools JimC Bioinformatics 7 05-21-2013 11:34 AM
samtools mpileup and bcftools not reporting dbSNP ids dmb Bioinformatics 2 08-13-2012 10:18 PM
samtools mpileup and bcftools command lines to compare individual differences? aslihan Bioinformatics 1 11-17-2011 02:20 PM
Conflict between mpileup/bcftools and GATK in VCF file ericarcher Bioinformatics 0 09-25-2011 04:33 PM
questions about samtools mpileup & bcftools chenjy Bioinformatics 0 07-26-2011 04:21 AM

Thread Tools
Old 08-05-2011, 09:04 AM   #1
Junior Member
Location: San Francisco

Join Date: Jun 2011
Posts: 6
Default Samtools mpileup/bcftools

Hi all,
I read through samtools manuals several times, but I'm still not clear on how exactly samtools & bcftools decide to call a SNP. I've tried to run through multiple combination of arguments with mpileup (-B, -C, -q, etc) & bcftools, but still ran into the problem below. I even ran bcftools view on the bcf file without the varFilter step, but the problem persists.

I have 2 samples, an original & an "evolved" cell line. Based on numerous runs, I found that there are many SNPs being called only in the "evolved" cell line but not on the original, making it look like they're "novel" SNP. However, when I view them on IGV, I can see the SNP in the original cell line and there don't seem to be significant differences between the mapping quality or base quality at the SNP position in these 2 samples.
It's not important to me if reads below a certain mapping quality don't get count, but the trouble is that it seems to be inconsistent. In one sample, SNPs on reads having mapping quality of 0 don't get count, but then they would get counted in the other sample, making it difficult to identify the true novel SNP. Is there anyway to force the SNP count to be more consistent?
cristae8 is offline   Reply With Quote
Old 04-11-2012, 08:38 AM   #2
Junior Member
Location: canada

Join Date: Apr 2012
Posts: 2


I'm also having a similar problem! I have PCR confirmed SNPs that I can see in IGV when I look at the BAM file, however they are not present after varFilter. I'm guessing that there are parameters that I need to change to include them, but there doesn't seem to be any trend in map quality or coverage that I can distinguish ones that are called vs. missed in the vcf output.

I mapped my PE Illumina reads with BWA and then running samtools mpileup:
$ samtools mpileup -uf x.fasta Initial-sorted.bam | bcftools/bcftools view -bvcg - > initial.raw.bcf

Then ran:
$bcftools view initial.raw.bcf | varFilter -D500 > initial_snps.vcf

I've tried adding -E switch to the samtools mpileup and get alot more SNPs, however this still does not include the SNPs confirmed by PCR. Also tried increasing the -D value when on the bcftools commands - again increases SNP count, but still not ones I know of.

Please help! Not sure what else to do....
tlynch1 is offline   Reply With Quote
Old 04-11-2012, 09:35 AM   #3
Location: Pittsburgh, PA

Join Date: Feb 2011
Posts: 49

Try VarScan

It takes the 'mpileup -f reference' output (piped or in a file) and then calls variants based on statistical analysis. Their paper does a good job at describing the algorithm. Plus the command line parameters are much more straightforward.

I hope this helps!
twaddlac is offline   Reply With Quote
Old 05-02-2012, 12:43 PM   #4
Junior Member
Location: canada

Join Date: Apr 2012
Posts: 2

Thanks so much! The VarScan results seemed to agree much better with my pcr data!
tlynch1 is offline   Reply With Quote

bcftools, mpileup, snp call

Thread Tools

Posting Rules
You may not post new threads
You may not post replies
You may not post attachments
You may not edit your posts

BB code is On
Smilies are On
[IMG] code is On
HTML code is Off

All times are GMT -8. The time now is 12:48 AM.

Powered by vBulletin® Version 3.8.9
Copyright ©2000 - 2020, vBulletin Solutions, Inc.
Single Sign On provided by vBSSO