Bioinformatics Tools to analyze diabetes susceptibility genes

[sojiadewumi]

Greetings. I am writing this to ask for answers in my academic research. I am working on using Bioinformatics Tools to analyse diabetes susceptibility genes/loci. I will appreciate if you can channel me in the right course with the view of knowing the right tool to use.

[gringer]

Type 1 Diabetes, or Type 2 Diabetes? They have very different heritabilites, and very different population prevalences, so the methods used to analyse them will also be quite different.

For Type 1 Diabetes, there's a huge association signal at the HLA region, so be prepared for analysis of a region of the genome that is extremely complex, and most programs will probably not be designed to handle this complexity.

[sojiadewumi]

Thanks, Gringer. That is much help. My research is based on Type 1. I wish you can provide further help as you said that most programs may not be able to handle it. Hoping to read from you soon. Many thanks.

[gringer]

Unfortunately I can't provide much further help because I only used Type 1 Diabetes in my research as a toy example for validating a population-subsampling genomic association method -- I didn't really care much about the trait, or how informative the results were. Just in case it's useful, you can find that here (T1D association study using WTCCC data is pp. 131-161).

I have noticed through a really quick search just now a couple of papers that might be useful:

• Original WTCCC study (including T1D)
• Follow-up meta-analysis (specifically for T1D)

With T1D, you're going to need genetic data from a lot of people because the population prevalence is so low. The LoL bioconductor package was used by the follow-up study, so that might be a reasonable place to start.

[sojiadewumi]

Thank you once again, Gringer. I have a leading to make you my mentor (Seriously). Your assistance has opened my eyes to some valuable information. What in your own estimation do you think can be feasible to work on with T1D. I wouldn't mind if you can give me some protocols. You can reach me via my email - [email protected]
Many thanks

[gringer]

[copied over from researchgate]

Hi, sorry, I got a little distracted with other work over the past week or so. The Wellcome Trust data can be accessed after application to the data access committee <[email protected]>, so that's unfortunately another application process that you'll need to go through:

http://www.wtccc.org.uk/ccc1/

It looks like the t1dgc data (possibly the same set of data) is similarly behind a data application wall (which I presume you already know):

https://www.t1dgc.org/views/vw_databases.cfm

And then there's the dbGaP route (which also requires an application):

http://www.ncbi.nlm.nih.gov/projects/gap/cgi-bin/study.cgi?study_id=phs000180.v2.p2

http://www.ncbi.nlm.nih.gov/projects/gap/cgi-bin/study.cgi?study_id=EGAS00000000038

You might also be able to get something out of the sequence read archive (which is a lot more publicly accessible), but that's probably only going to be a few individuals (although *very* high coverage), rather than the thousand or so that are included in the WTCCC dataset(s):

http://www.ncbi.nlm.nih.gov/sra/

Sorry I can't help too much more on that front except to say, "hang in there".
---

Regarding other protocols, there's GenABEL and plink for association analysis. If you want to get into haplotype analysis (which I think would be a good idea for complex genetics over a small region), then Haploview allows you to do haplotype-based association tests, and can give you a better idea of the recombination background of your regions -- might struggle a bit with WTCCC data though.