Forum:Is this convention canonical for upstream variants?

Question

[Ram]

I am looking at this variant: http://www.ncbi.nlm.nih.gov/projects/SNP/snp_ref.cgi?rs=754264214 and it is represented as NM_000309.3:c.-8-11C>T

The gene PPOX (NM_000309.3) has a 5' UTR, of which 8 bases lie in first CDS exon. Usually, upstream variants are denoted by a c.-[number_of_bases][REF]>[ALT], but here I see c.-8-11, which could indicate that the variant is 19 upstream of CDS start, and 11 upstream of CDS_Exon-1.

Is this representation canon/acceptable? Can we insert random pieces of information into standardized nomenclature for supposed clarification?

P.S: Let us ignore the part where an NM identifier is followed by a negative co-ordinate - I am not sure if that's valid, but I see it quite often. The problem of course, is that NM sequences may not always have the REF base at the upstream position - they may only include the sequence from a later position.

[Devon Ryan]

That nomenclature makes absolutely no sense to me, though people seem really good at coming up with the craziest HGVS representations.

[Ram]

But does dbSNP not curate entries? I mean, expecting them to follow the latest in HGVS rules might be problematic, but this?

Answer 1

[Ram]

A day wiser me has reached a different conclusion: This nomenclature makes sense.

NM_000309.3 (PPOX) is a gene where an entire exon is part of the 5' UTR. Denoting c.-8-11C>T as c.-19C>T would not signify the fact that this is an intronic variant, not a regular upstream variant. By using c.-8-11C>T, the notation tells us that this is a variant located upstream of the start codon, in an intronic region that falls inside of the 5' UTR.

[Devon Ryan]

Ah, that makes some sense. I'm again glad that I get to mostly avoid dealing with variant nomenclature :)