Fewer reads in the middle of some exons

[rkawa]

Hi everyone,

I have encountered somewhat unique issue in one of my collaborators samples and would like to hear your input/thoughts.

1. Reads in the middle of some exons are lower in one set of samples (triplicates).
2. This happens pretty randomly in many exons but not in UTRs.
3. It is not observed in short exons.
4. Library was made at the same time with other samples.
5. All samples were spread in several lanes to avoid batch effect during sequencing.
6. Reads were aligned with STAR
7. Seems to appear too often to consider alternative splicing/differential exon usage
8. In the example attached, the bottom two have fewer reads in the middle of snap25 exon.

Thanks for reading and appreciate your help!

[Brian Bushnell]

Have you tried a different aligner?

Could this be a repetitive element or highly-variable region that might be misassembled or have different copy count / structure in different individuals? A single SNP could change the mapping from one location to a highly similar location. Also, what is the mapping quality of reads in these areas compared to other areas?

Is there a known pseudogene related to this gene?

[rkawa]

Hi Brian,
Thanks for quick reply. I haven't tried different aligner. SNP is unlikely because these are all mouse samples from the same genetic background. Mapping quality of the reads are generally good (uniquely mapped reads ~80%, unmapped ~0%). There aren't any mismatches in the regions in question. Happens fairly frequently in many genes in this sample but not the others. It is not repetitive sequences which have this phenomena.

Thanks

Riki