Hello everybody I'm running snpcalling on RNASeq data. My data are related to 6 sample including two conditions, control and treatment with 3 replicates for each. I want to use the star aligner. I am confused at alignment step. Do I need to align three replicate of each condition with genome reference in one run and create one BAM file for each condition? then, I assign reads of each BAM file (3 replicate of control or treatment) to a specific read group with AddOrReplaceReadGroups (picardtools). So, I will achieve one vcf file for each condition and then I have to compare them.
OR
DO i need to align each replicate (Including 2 fastq file) with my reference genome and create a BAM file for each replicate. then, should I assign reads of each BAM files (each replicate) to a specific reads group with AddOrReplaceReadGroups. next step, should I merge vcf files related to the replicate of each condition. Eventually I get two vcf files, each of which is the result of merging the vcfs of replicates.
Thank you
•
written
by
eyonesi •
You should map individual sample and then make a combined vcf file. Its always good to make merged vcf file using from bam files rather than making individual vcf files and then merging it.
I'm not sure if this is good advice. If the replicates are from the same individual, combining them will increase the coverage and as such be helpful for calling variants. Especially in RNA-seq the coverage distribution is very uneven, and for lowly abundant genes this can make a big difference to have 3x the coverage.
These samples do not come from the same individual? So you expect them to have different variants? In that case just align everything separately.
Log in to answer this question.
Are those technical replicates, originating from the same individual or cell line?