Ucsc Browser And Novel Snps

Question

[User 6659]

Hello

Can the UCSC browser be used to predict the effect of novel SNPs on a transcript (e.g. synonymous and non synonymous)? I was wondering if UCSC did any 'processing' on novel snps (e.g. looked to see what transcripts they might be in and find the consequences) or whether it looked for intersection with existing SNPs in the database and just reported the consequences of those SNPs already in the database. Or perhaps it does neither of those.

I have seen the SNP analysis tool ANNOVAR claim that it can annotate novel SNPs but this tool is based on the UCSC browser tracks. I was wondering if the ability to handle novel SNPs is written into the UCSC browser or whether ANNOVAR has added an additional annotation layer to supplement the UCSC information.

thanks

Answer 1

[brentp]

Annovar downloads text file versions of the data you see in the UCSC. The reason it is novel (compared to UCSC) is that it adds information to your SNP calls en masse. Annovar can use knownGene and refGene annotations to determine the effect of a SNP in an exon and it can use dbsnp131 (or whatever version you choose) to tell you if it's a known SNP. You could potentially view each SNP call that you have individually in the UCSC browser, but Annovar is one way to automate this type of exploration.

[User 6659]

Hi Brent. That wasn't really what I was asking. Or perhaps i didn't word it very well. If you give a SNP to UCSC or annovar that doesn't exist in dbSNP, how do they know the effects of the SNP? They can easily see if the SNP is in an exon by looking at genomic coordinates but how do they know if the SNP is synonymous/non synonymous etc. Do they actually do the necessary analysis to work this out. If the UCSC does this, where is the code stored to do it?

Answer 2

[User 6659]

I emailed the author of the tool and he confirmed that the tool uses genomic sequences to calculate the effects of novel variants

[husainix]

Hi User 6659, Im actually finding on information on this as well just in case you've solved it. How do they determine whether the novel snps (not in database) are synonymous/nonsynonymous? and how can we predict and view the amino acid change? if so, on what basis, any software or algorithm I should understand? and how can we see the effect on protein functions, stability, interactions.

Im a postgrad student. just starting my MSc. Im very new in this area, basically doing this on my own, perhaps need validation, and hope to learn a lot from you guys :)

A summary on what ive been doing. Correct me if im wrong ya. What I already did is using annovar to filter the .vcf of human genome im studying against dbSNP and get the filtered SNPs which we can say, novel.

And from here, I need to work on with this novel SNPs. From my understanding, I filter this with avsift so that I can get the nsSNPs of this novel SNPs, get the SIFT score, and determine whether those novel SNPs are tolerated or intolerated. So those which are intolerated may hv impact on protein functions. Next, I will narrow down my research to study those deleterious nsSNPs.

So, what do you think about this. Is what im doing here make sense?

Thank you. Cheers