Question: Best way to link drugs to diseases preferably using R
gravatar for bioinformatics
2.5 years ago by
bioinformatics0 wrote:

Hello !

We have a list of drugs that reverse a particular biological phenotype. Now we want to correlate each drug to diseases to approximate the danger of these substances.

The end result should be something like a list of pvalues(or distances..) for each drug to promote particular diseases.

Does anyone know good ways to calculate correlations from drugs to diseases?

​We know about DrugVsDisease, but is there something similar or better ?

Not sure if what I am mentioning here can already somehow be achieved using the DVD package. In my own experience I think it can only be used to do drug-drug and disease-disease relationships. If this is the way to go please let me know !

We only have the drug name, so the diseaseRL data(differentially expressed genes) from the DVD package may be useful.

We would like to use R, so it is possible to select on different characteristics.

​Best Regards, Rick

R disease drugvsdisease drug • 874 views
ADD COMMENTlink modified 2.4 years ago by Egon Willighagen5.2k • written 2.5 years ago by bioinformatics0

Take a look at the Open targets platform in addition to other suggestions.

ADD REPLYlink written 2.4 years ago by genomax75k

Check our blog post How to access Open Targets with R. This R client is no longer maintained by the team but feel free to contribute to it (and pull requests please).

ADD REPLYlink written 2.3 years ago by Denise - Open Targets5.0k

Does anyone know good ways to calculate correlations from diseases and drugs?

I think you'll have to give some more specifics about what your end game is here. Let us know if you're looking to explore the relationship between disease, gene expression and drug response, or if you have genes from a disease and want to know what drugs target their expression, etc.

ADD REPLYlink written 2.5 years ago by andrew.j.skelton735.9k

Dear Andrew,

Thank you for your quick response. I have modified the context of the question, and hope that it is more clear why and what we would like to achieve.

So to summarize:

We have list of dugs. We want link each drug to diseases ( side effects )

So end result :

each drug linked to diseases(side effects) with pvalues(or maybe some other measurement) for each disease.

Example :


disease              pvalue

Hair loss             0.7
cancer                0.3

The diseases here are just examples, and arbitrarily chosen.

Regards, Rick

ADD REPLYlink modified 2.5 years ago • written 2.5 years ago by bioinformatics0

To do this, you need to have data linking diseases and drugs even in an indirect way and how to proceed depends on the data you have. So what kind of data do you have ? If your question is about what data to use for this, maybe you can make use of the database of side effects (SIDER)

ADD REPLYlink written 2.4 years ago by Jean-Karim Heriche21k
Please log in to add an answer.


Use of this site constitutes acceptance of our User Agreement and Privacy Policy.
Powered by Biostar version 2.3.0
Traffic: 1049 users visited in the last hour