Which Human Sequence Should I Take To Alter With Snp Variants??

Question

[Cynthia]

Hi Friends,

i have taken reference sequences for my comparative genomics analysis with other primates. And i have collected the SNPs which mainly causes the disease.

Now i have to check out whether these SNPs are present in human as well as in other primates also, for the same purpose i need to have one sequence(human) which will consists of all the SNPs which i ve found responsible for the disease.

And compare wid the sequence consists of SNP variant with my reference sequences which i ve taken before.So, how can i mutate the sequence? which sequence(human) should i take to alter with SNP variants??

[Larry_Parnell]

The above comment makes this even more confusing. Could be a good questions if edited for clarity.

[Cynthia]

the nucleotide sequence or the reference sequence shud i choose to add the mutations?

Answer 1

[Sean Davis]

In addition to the comments by Jorge, you might consider also using the UCSC liftover tool to remap your various model-organism SNVs over to human. From there, it is a simple matter of matching on location, now all with respect to the human genome.

[Jorge Amigo]

I didn't know that you could use liftover to map things between different organisms, but if that is the case that would be a more straightforward option, at least for all the variants that would be covered by liftover (which may not be all).

Answer 2

[Jorge Amigo]

I would suggest you to rewrite the question, since its meaning is not clear. just in case I understood you right, let me throw some ideas in advance:

1. do you need a human sequence that matches your other sequences? then you will need to do a blast search with your other sequence, that ideally would give you a start and end position of the human chromosome.
2. once you know that start and end positions, I would query dbSNP and see if there are any SNPs reported on that region.

[Cynthia]

Actually i ve found out the SNPs which are resposible for causing the disease. Now i ve to check whether these SNPs(which i ve found out through literature search) is present in my sequence or not.

for eg: rs6094752 is one of the SNP in the gene NCOA3 causing Breast cancer.

and i have a referce sequence of human NM_001174087. now i have to check whether rs6094752 is present in NM_001174087 or not? this is my question. How can i find it?

[Cynthia]

Actually i ve found out the SNPs which are resposible for causing the disease. Now i ve to check whether these SNPs(which i ve found out through literature search) is present in my sequence or not.

for eg: rs6094752 is one of the SNP in the gene NCOA3 causing Breast cancer. and i have a referce sequence of human NM_001174087. now i have to check whether rs6094752 is present in NM_001174087 or not? this is my question. How can i find it

[Cynthia]

I guess i'll be clear now if i give an example.

rs118166747 is the SNP i've got it from OMIM, Now i need to check rs118166747 is present in other primates or not?

[Jorge Amigo]

the rs codes will not help you, since they are organism dependent. since you will then need to query that particular site on several organisms I would definitely use a blast search (using the flanking sequence) or UCSC's liftover tool as suggested by Sean. once you know which base position corresponds on each organism to that human position (if it applies, of course), you may then query variation repositories (Ensembl, dbSNP, ...) with those positions obtained previously. but as Larry says, this search is very unlikely to work, and if it does it may return spurious results, so be careful.

Answer 3

[Larry_Parnell]

Basically, as I've stated before, a SNP in one primate is highly unlikely to be variant in another primate. It is more informative to look at ancestral alleles across your multiple sequence alignment.

A point of note: typically, a SNP is not known to cause a disease, but to be genetically linked to a change in disease risk. This is an important distinction especially as functional data on most risk-associated SNPs is lacking.

[Jorge Amigo]

I guess that with your note you are referring to rare diseases, since there are indeed SNPs that are on their own associated with a diagnosis. anyway, I totally agree that thinking carefully if a query should be performed before performing it is always advisable, since the results you may obtain from a poorly designed query can be spurious, and may lead you to wrong conclusions.