I am combing through my 23 & me raw data and I am a little confused on SNP terminology. I am using NCBI's genome browser and SNP database. As an example we can all follow here is a link to a CYP1A1 SNP https://www.ncbi.nlm.nih.gov/snp/rs28399430. When you open this link you will see alleles listed as G>C. Which one of these is the reference genome and which is the pathogenic allele? Is the relationship always constant (like in x>y, x is always the reference and y is the substitution). If you scroll down to the genome browser view, there also doesn't appear to be any way to tell + vs - sense strand (ie making it difficult to tell G for C mutations or vice versa). I would just assume its the top strand? The ultimate goal here is for me to look through 23&me and compare my SNP genotype to the reference genome and see if my genotype exhibits pathogenic SNPs in certain genes.

  • $\begingroup$ > "Which one of these is the reference genome and which is the pathogenic allele?" How have you concluded that the less common, non-reference allele is pathogenic? I don't see any clinical data for that $\endgroup$ – swbarnes2 Apr 10 at 20:37
  • $\begingroup$ For specific examples I’m looking at it’s the case. Not globally- should have made my self clear sorry $\endgroup$ – Joe Apr 10 at 22:28

Quick answer is yeah, the G is the "reference allele" as you put it. The greater than symbol is acting as an arrow, ie, G goes to C, that is, G is replaced with C.

Actually you get a pretty clear indication of that if you look down a line it two where it mentions the frequency:

C=0.000419 (104/247982, GnomAD_exome) C=0.000494 (62/125568, TOPMED) C=0.000538 (62/115296, ExAC)

The C frequency is way too low to be the "normal" allele. It's rather the substitution which results in a missense variant. You can see more specific stats by clicking on the frequency button. In this case they basically just say that in all studied populations the G variant is by far the most common.

| improve this answer | |

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service, privacy policy and cookie policy

Not the answer you're looking for? Browse other questions tagged or ask your own question.