I recently analyzed some
targeted exome sequencing samples, which were provided to us by our collaborators, for which
I do not possess the target gene list. Upon analysis, I am informed that some of the
genes - whose variants were identified - were not present in the target gene list. Has anyone ever faced such an issue, or have any idea why I might be observing these variants?
If it helps, I had found duplicate entries (both name and sequence) in some raw fastq files, so I had removed them using
seqkit rmdup. Since I don't know whether all variants in untargeted genes exist exclusively in these files, I can't even be sure that removing the duplicate entries could be causing an issue with the alignment and/or variant calling.
The pipeline used was -
trim_galore (while preserving only the paired reads, and not singular reads) -->
seqkit rmdup -n (to remove duplicate entries based on name) -->
bwa_mem using hg38 as the reference -->
sort_sam, mark and remove PCR duplicates --> variant calling with
GATK4 --> BQSR with
GATK4 --> applying BQSR on bam file with
GATK4 --> variant calling from the recalibrated bam file created in the previous step using
GATK4 --> annotation using hg38 as a reference with wANNOVAR
Thanks in advance.