This is a test version of Biostars. For the public version, visit https://www.biostars.org.
Gene Disease (Cancer) Association

I'm a computer scientist, so I don't have a lot of knowledge about biology. I developed a statistical method to highlight some genes from microarray dataset. Now, I'd like to know if the gene I selected are related to the microarray problem. That is, given a microarray dataset about Prostate Cancer (or Leukemia), I'd like to know if genes: X,Y,Z, etc. are already known to be related, in literature, with Prostate cancer. Is there any tool/website that could help me? Is there a way that, given a gene, I know all the disease associated to it? Or, in the other way around, what's the better method to get all the genes associated with a particular disease?

UPDATE:

I'd suggest an interesting tool that I've found: Diseases. It mines the literature to associate genes to diseases or viceversa.

gene association cancer disease function
• 4,799 views • link
written by TitoPullo TitoPullo

3 answers

That's exactly what you will find if you download data from the The cancer Gene Census at http://cancer.sanger.ac.uk/cancergenome/projects/census/

• 242 views • link
written by Arnaud Ceol Arnaud Ceol

But this website list all the studies involving that particular gene. It doesn't tell me if that gene is actually related with the disease.

• 0 views • link
written by TitoPullo TitoPullo

"The cancer Gene Census is an ongoing effort to catalogue those genes for which mutations have been causally implicated in cancer.": you can download the Excel or tab delimited file, it lists all genes and associated cancer types.

e.g.: Symbol Name GeneID Chr Chr Band Cancer Somatic Mut Cancer Germline Mut Tumour Types (Somatic Mutations) ABI1 abl-interactor 1 10006 10 10p11.2 yes AML

• 0 views • link
written by Arnaud Ceol Arnaud Ceol

There's some good sources already in a comment, but I'd like to highlight a few more:

  • PubMed: Just search for the gene, really, and look at studies. If you're into the more advanced stuff, look at co-mentions (i.e. in how many articles are the gene and the word prostate cancer, or any derivative thereof, co-mentioned)
  • ExpressionAtlas: Look at other peoples results, and compare.

I would just like to point out an obvious problem with what you are asking: "to be related to disease X" can have multiple meanings. Related how? A mutation is causing the cancer, it's a prognostic marker of the cancer, overexpression drives the cancer and so on.

Also, be aware that there are a lot of different statistical methods for microarray analysis. If you ever submit a paper using this method, be aware that a lot of reviewers will be very critical about methods they're not familiar with (this is my own opinion, and I only have my own experiences to back it up, and is not meant to be discouraging.)

• 215 views • link
updated by Ram Ram • written by David Westergaard David Westergaard

I think PubMed is the best resource for me.

• 0 views • link
written by TitoPullo TitoPullo

Along with the ones already mentioned you could search the OMIM genemap for genes with phenotypes associated with a specific disease.

• 0 views • link
written by Vivek Vivek

Log in to answer this question.

More posts like this