What is the advantage of having BAM files per sample compared to one BAM file containing all samples?
Hi,
I have multiple BAM files per sample and I would like to call SNPs from all of these files to analyze all samples together afterwards. Could it improve the SNP calling, when merging the BAM files together? Or does it make a difference for the VCFs produced afterwards?
Hope you can help me.
Best regards!
• 1,833 views
•
link
written
by
L_to_the_m •
0 answers
No answers yet.
Log in to answer this question.
More posts like this
-
Generating vcf files per individual instead of one big file from bam files
written by salman_96 •
-
Imputation with Beagle 5.2 for whole exome data at once
written by L_to_the_m •
-
Pipeline for familial sequencing
written by saliksyed88 •
-
How does samtools mpileup handle multiple .bam inputs?
written by traviata •
-
Samtools: merge and mpileup vs mpileup alone for variant-calling with multiple BAM
written by Rob
-
Combine single sample vcfs together
written by sumudu_rangika •
-
Merge and fixmate of multiple BAM files generated from pair-end DNA sequences
written by lisadavic66 •
-
How can I filter SNP for 30 samples in vcf files?
written by sun.nation
-
Snp Calling Of Tcga Data Using Samtools
written by pm2013 •
-
Gatk Multi-Sample Calling
written by Bioscientist
pooling of BAM files was performed for 1000 Genomes project since the coverage of the whole-genome sequencing was low.
it allowed the detection of frequent SNPs with good accuracy.
nowadays, having a coverage of >30x on average, it does not make much sense to pool BAMs - even rare SNVs can be quite confidenlty detected (more complex with indels + complex regions, but this is another topic).
so the answer is - it depends on your task, do you want to find common SNPs in low coverage samples or you care about rare variants in OK covered ones.
Thanks for your reply. I want to analyze them later on all together in one PLINK file. So I want SNPs for all samples with high hard call rate. So I am not sure, if I understood it correctly what is the best way to do this?
you can perform multi-sample calling using GATK - https://gatk.broadinstitute.org/hc/en-us/articles/360035890431-The-logic-of-joint-calling-for-germline-short-variants - but I doubt you need to physically merge reads from different samples (thus merging their "BAMs")
Ok thanks, so the link you shared indeed talks about the advantages of joint genotype calling. That makes much more sense than just merging BAMs. Thanks a lot!