Include splicing terms as children of coding_transcript_intron_variant

[tedil]

Right now, the following splice related terms

• splice_acceptor_variant
• splice_donor_5th_base_variant
• splice_donor_variant
• (and their children)

are children of splice_site_variant, which in turn is a child of intron_variant.
However, right now, they are not children of (non)coding_transcript_intron_variant, which is a bit unfortunate because according to (at least my understanding of) the definitions they should be (i.e. in addition to being a child of intron_variant). This makes programmatic checks more cumbersome than they need be.

Additionally, the terms:

• splice_polypyrimidine_tract_variant
• splice_donor_region

are technically also intron_variants, but do not seem to be on any path that goes through intron_variant (or (non)coding_transcript_intron_variant, for that matter).

(splice_region_variant does not fit here, because that also includes exonic_splice_region_variant, which is clearly not an intron_variant)

Could these additional relationships be added? Or did I get something wrong?

edit: While at it, should splice_donor_5th_base_variant also be a child of splice_donor_region_variant? Or does this have to do with the classification into *_site and *_region variant?

[egchristensen]

@tedil If we added coding_transcript_intron_variant SO:0001969 and non_coding_transcript_intron_variant SO:0001970 as parent classes of splice_site_variant SO:0001629, should intron_variant SO:0001627 also be kept as a direct parent class or should it be removed? Either would be valid from an ontological perspective, but I'm not sure if removing it will have implications for your programmatic checks.

While at it, should splice_donor_5th_base_variant also be a child of splice_donor_region_variant? Or does this have to do with the classification into *_site and *_region variant?

I'm not sure. Since the definition of splice_site_variant explicitly mentions changes to the 5th base pair after the start of an intron and the definition of splice_region_variant SO:0001630 excludes any changes more than 3 bases away from an exon I'm inclined to say that it should stay as is. I guess the answer might have something to do with the number of bases selected for use in these definitions. What do you think?

[tedil]

@tedil If we added coding_transcript_intron_variant SO:0001969 and non_coding_transcript_intron_variant SO:0001970 as parent classes of splice_site_variant SO:0001629, should intron_variant SO:0001627 also be kept as a direct parent class or should it be removed? Either would be valid from an ontological perspective, but I'm not sure if removing it will have implications for your programmatic checks.

Good question. I wouldn't want for you to cater to my specific programmatic needs, but rather on what needs to change to properly and correctly encode the respective relationships within the SO. (Removing it has no implications for me personally btw, but should probably be kept to not unnecessarily introduce more changes than necessary? But then again you'd probably want to avoid cycles…)

While at it, should splice_donor_5th_base_variant also be a child of splice_donor_region_variant? Or does this have to do with the classification into *_site and *_region variant?

I'm not sure. Since the definition of splice_site_variant explicitly mentions changes to the 5th base pair after the start of an intron and the definition of splice_region_variant SO:0001630 excludes any changes more than 3 bases away from an exon I'm inclined to say that it should stay as is. I guess the answer might have something to do with the number of bases selected for use in these definitions. What do you think?

splice_site_variant:

A sequence variant that changes the first two or last two bases of an intron, or the 5th base from the start of the intron in the orientation of the transcript.

splice_region_variant:

A sequence variant in which a change has occurred within the region of the splice site, either within 1-3 bases of the exon or 3-8 bases of the intron.

splice_donor_region_variant:

A sequence variant that falls in the region between the 3rd and 6th base after splice junction (5' end of intron).

splice_donor_variant:

A splice variant that changes the 2 base pair region at the 5' end of an intron.

splice_donor_5th_base_variant:

A sequence variant that causes a change at the 5th base pair after the start of the intron in the orientation of the transcript.

From those definitions, I would have expected that splice_donor_5th_base_variant is (additionally) a child of splice_region_variant, because it clearly falls within the 3-8 bases of the intron part of the definition of splice_region_variant.

What about splice_polypyrimidine_tract_variant and splice_donor_region? These should be descendants of intron_variant in some way or another, right?

Thank you for looking into this!

[p1k2r3-code]

There does not need to be a separate entry for splice_donor_5th_base_variant, since this position has information content above background levels that makes it part of the 5' or donor splice site. See Caminsky et al. 2015 ((https://doi.org/10.12688/f1000research.5654.2). There are many published disease mutations involving this nucleotide location.

Information weight matrices for acceptor splice sites contain positions -25 to +2 (including a zero position just upstream of the splice junction) and for donor sites -3 to +6 (including a zero position just downstream of the splice junction)