Searching for human genotyping data thats not public

Question

[Bioinformatics-man]

I have a request from an investigator to acquire genotyping data.

He is asking for genotyping data of any given human population. He wants allele and frequency data for most clinical variants of a few genes of interest.

So, I used the dbSNP database for the files. I sent them, but he replied explaining that the data needs to be novel and not public.

'Novel' meaning, it hasn’t been used in publications? I don’t see the point in acquiring some external data without any more information provided. Why would this data be any less useful?

Also, can I select non public options with dbSNP? If not, where would I acquire “novel” genotyping data?

[4galaxy77]

I think you should tell the investigator that he should fork out the money to newly genotype individuals if he wants novel data. Not sure there's any other way of getting 'novel' data.

Answer 1

[predeus]

That's a very strange request. You really need to clarify it with him - as in, what is the idea behind this.

The best place for human allele frequencies is gnomAD; however, these variants can hardly be viewed as novel. By far not all of them are characterised, of course.

If he's looking for smaller populations with some unique variants, that's possible, but certainly would not be available without a publication. E.g. we published allele frequencies from of Northwest Russia, with approximately 700 people in the larger cohort. Obviously there's lots of similarity with general Europeans and Finns, but also some unique components.

Finally, you can get access to some protected studies using dbGap. It's quite hard though, and you need to know exactly what you want before going through with the application process.

[Bioinformatics-man]

Hey, thanks for the advice!

I looked into dbGap but I noticed that a PI would have to submit a request anyway.

I will update him and see what he wants.