In analysis of human genomic variants based on either WGS or WES, I often came across high impact variants shown as deletion of introns at their entirety. Since there is no known molecular mechanism, which specifically removes the introns at the genomic level, I can only think of two possible scenarios for this to occur.
The 1st possibility is contamination of mRNA in the preparation of genomic DNA samples. However, I can't think of a way for the mRNAs to survive the DNA sequencing library preparation steps.
The 2nd possibility would be the existence of novel processed pseudogenes. In this case, the reads representing the processed pseudogenes would be mapped to their parent genes as discordant reads with deletion of introns due to the absence of introns in their sequences.
I wonder if anyone else has had similar observations and has an ideal about the reason(s) behind.
Thank you very much!
Ping
Just to add: Such deletions tend to happen in the same set of samples, suggesting to me that the RNA contamination is likely to be responsible. Also, I'd think it is important to understand the situation, as such deletion variants happen very often and can be very distractive as false positives.
I'm not sure what you mean by 'high impact deletions'. Do you have an example?
I'm also not sure why you wouldn't have deletions of introns. A deletion of an intron is unlikely to have severe consequences to the gene.
Hi Christian,
Thanks for looking into my question. When an entire intron is detected as deleted, it removes the canonical splicing signal, so it is usually considered to have high impact. I guess the impact here doesn't really matter here. The essence of the issue here is why the deletion of introns occur. Is it an artifact or is it real? Looks to me it is more likely to be an artifact. But I don't what causes it.
Ping
Below is an example (a partial vcf entry, genomic coordinates based on hg19):
chr2 172644171 . GCTGAAAAAGAGAAGCAGGGGCAGGGGAGACTTGAAACCAGGACAAATGTGGTAAAGATAGCCACTGAGTCACAGGGGAGGACGCTAGAGCCAGCCAGCCATTCTGTATGGCTCCAGCCCCTGCCTAC G 84.03 VQSRTrancheINDEL99.00to99.90
Sure the canonical splice sites are removed, but introns are removed from mature transcripts anyway. It's probable that the loss of an intron from the DNA is unlikely to have a detrimental effect on the protein function.
In terms of whether it's real or artifact, have you tried to validate via PCR and Sanger sequence?
Hi Christian: Validation has not been done for all cases. But in one case, Sanger sequencing failed to detect the deletion, which would be what we can expect for processed pseudogene or mRNA contamination. So I guess you haven't seen similar situations in your mutation analysis.
Depending on your pipeline for discovery INDELs can be more problematic and generate false-positives.
I can't say I've seen similar situations, but in our data we tend to focus on protein affecting changes so 'silent' intron affects wouldn't be obvious to us.
- Badges
- Science topic
- Similar topics
- Biological Science
- Bioecology
- Systems Ecology
- Landscape Ecology
- Connectivity