How to get SNP id (rsid) from this input "myGene 1234C>A" ?

Question

[zx8754]

Came across this paper recently:

HSD3B1(1245A>C) variant regulates dueling abiraterone metabolite effects in prostate cancer. - Alyamani M, et.al.

https://www.ncbi.nlm.nih.gov/pubmed/29939161

It mentions HSD3B1 1245A>C variant.

First, I assumed 1245 refers to base pair position on the chromosome 1, but the gene position is chr1:120049825-120057681 using UCSC (http://genome-euro.ucsc.edu/ ):

#hg19.knownGene.name    hg19.knownGene.chrom    hg19.knownGene.txStart  hg19.knownGene.txEnd    hg19.kgXref.geneSymbol
uc001ehv.1  chr1    120049825   120057681   HSD3B1
uc031pnt.1  chr1    120049940   120057681   HSD3B1

Searching for chr1:1240-1250 gives nothing.

Question:

1. What is the rsid for this variant?
2. What are your usual steps to identify this type of ids?

Answer 1

[Emily]

1. I don't know the rsID of this variant.

2. I would assume the authors have attempted (and failed) to write in HGVS nomenclature. To write correct HGVS describing the position of the variant in a CDS you write it like:

versioned transcript identifier:c.position in the CDS of the variant reference base > alternative base

eg ENST00000369413.8:c.1107G>A

Unfortunately, many people, like the authors here, are idiots, and don't realise that without a versioned transcript identifier, it is completely ambiguous as to the base they're referring to. Fortunately, the Ensembl VEP can read ambiguous HGVS assuming the coordinates are correct.

Unfortunately, in this case the coordinates are not correct, as they are talking about position 1245 in a 1122 base CDS. I assume they did this by cDNA rather than CDS, which is wrong. While coding sequences are relatively stable, UTRs are less so, so without a versioned transcript identifier, it's nigh on impossible to find out what they were talking about.

My solution would be to email the authors asking them for an rsID, genomic coordinates or at least proper HGVS, and email the journal asking what on earth possessed them to publish such rubbish,

Answer 2

[GenoMax]

All currently annotated SNP for this gene can be found by using EntrezDirect (truncated for brevity):

$ esearch -db snp -query "HSD3B1" | esummary |  xtract -pattern DocumentSummary -element SNP_ID,DOCSUM | head -4
1557897965      HGVS=NC_000001.11:g.119515445T>C,NC_000001.10:g.120058068T>C,NG_050909.1:g.13334T>C|SEQ=[T/C]|GENE=HSD3B1:3283
1557897925      HGVS=NC_000001.11:g.119515378G>C,NC_000001.10:g.120058001G>C,NG_050909.1:g.13267G>C|SEQ=[G/C]|GENE=HSD3B1:3283
1557897875      HGVS=NC_000001.11:g.119515294T>A,NC_000001.10:g.120057917T>A,NG_050909.1:g.13183T>A|SEQ=[T/A]|GENE=HSD3B1:3283
1557897867      HGVS=NC_000001.11:g.119515256C>T,NC_000001.10:g.120057879C>T,NG_050909.1:g.13145C>T|SEQ=[C/T]|GENE=HSD3B1:3283

Digging into publications that refer to this SNP I discovered this paper which mentions:

A single nucleotide polymorphism (SNP; rs1047303, NM_000862.2) in the HSD3B1 gene, a nonsynonymous variant c.1245A > C allele change, renders p.N367T and 3βHSD1 to be resistant to ubiquitination for proteasomal degradation, resulting in nearly 30% gain of function and subsequently enhanced synthesis of testosterone and DHT.

Using that RS number we currently get:

$ esearch -db snp -query "rs1047303" | esummary |  xtract -pattern DocumentSummary -element SNP_ID,DOCSUM 
1047303 HGVS=NC_000001.11:g.119514623C>A,NC_000001.10:g.120057246C>A,NG_050909.1:g.12512C>A,NM_000862.3:c.1100C>A,NM_000862.2:c.1100C>A,NM_001328615.1:c.1100C>A,NP_000853.1:p.Thr367Asn,NP_001315544.1:p.Thr367Asn|SEQ=[C/A]|GENE=HSD3B1:3283

Answer 3

[Ram]

I can answer #2.

I usually search by gene name on dbSNP and sort by chromosomal position. I then eyeball coordinates or use browser search to narrow down on the region. I prefer this way because it helps when people make typos in their coordinates (which was a major problem when I started doing this sort of search).