Skip to content
/ SIDER2 Public

Calculating compound similarities using side effects and indications extracted from drug labels

License

Notifications You must be signed in to change notification settings

dhimmel/SIDER2

Repository files navigation

SIDER 2

This repository contains an analysis of the SIDER 2 (side effect resource) data. This resource automatically extracted concepts from drug labels to create a database of drug side effects and indications.

This analysis provides the source for a companion website, which displays the analysis.

This project is divided into three parts and each part contains a single notebook:

  1. ./parse -- processing the SIDER 2 data in R
  2. ./similarity -- calculating compound-compound similarities in R
  3. ./compounds -- querying pubchem for compound attributes in python

The SIDER database was described in:

Kuhn M, Campillos M, Letunic I, Jensen LJ, Bork P (2010) A side effect resource to capture phenotypic effects of drugs. Molecular Systems Biology. doi:10.1038/msb.2009.98

To compute compound-compound similarity, we use an analysis similar to:

Campillos M, Kuhn M, Gavin AC, Jensen LJ, Bork P (2008). Drug Target Identification Using Side-Effect Similarity. Science doi:10.1126/science.1158140

About

Calculating compound similarities using side effects and indications extracted from drug labels

Topics

Resources

License

Stars

Watchers

Forks

Releases

No releases published

Packages

No packages published