Indirect targets by merging Drug-Target and Protein-protein databases

Question

[kakukeshi]

Is it possible and reliable to merge drug-target actions (inhibition or activation) with protein-protein actions in order to obtain the effect of drugs into indirect targets? The databases that I want to use are STITCH and STRING

Thanks

Answer 1

[Kevin Blighe]

I'm sure that it's possible by working with the data available to you. As to the reliability and interpretability of the results, that depends on the quality of the data that you're using. Both STITCH and STRING are hosted at EMBL, which is a reputable organisation.

If this is just a research project, then the results would be quite interesting. You could even attempt to bring in genomic information through the data available at The Pharmacogenomics Knowledge Base (https://www.pharmgkb.org/), although they may not cover the specific drugs/proteins in which you're interested.

If this is for some commercial or clinical purpose, then I would really question the results without further validation that proves what the results show.