Compared to other tissues, more exonic elements are found to regulate alternative cassette exons in testes.
In total, 637,841 variants were given exonic annotations by either ANNOVAR or VEP (Table 2).
There are 590,893 variants given exonic annotations by ANNOVAR using REFSEQ or ENSEMBL (or both), of which 488,113 (83%) had precisely matching annotations when using the two different transcript sets (Table 1).
This difference is much larger than the one when reads were generated from both exonic and non-exonic regions, indicating that a lot more exonic reads were incorrectly aligned by the direct mapping strategy due to the higher sequence similarity between human and mouse exons than other genomic regions.
A normalised value of 1.0 represents the detection of both exonic alleles (that is, no alteration), a value ≤0.65 is the threshold suggestive of loss of one exonic allele (that is, deletion) and a value ≥1.3 suggests the gain of one or more exonic alleles (that is, duplication).
Here we compare annotations from ANNOVAR and VEP using the ENSEMBL transcript set, focusing on exonic annotation categories.
The agreement is not as good for synonymous and splicing variants, but we observe that variants given an exonic annotation when using REFSEQ usually get the same annotation when using ENSEMBL.
They are more likely to be involved in disease and have more BP annotations.
More specific annotations are preferred over more general ones where there are overlapping annotations.
Interestingly, of all the exonic regulators in cerebellum and testes, only one contributes more exon inclusion in testes than cerebellum, indicating that a large set of exonic cis-acting elements are regulating exon skipping in testes.
