Adverse drug reactions (ADRs) are responsible for the failure of significant portion of investigative drugs and are the major reason for the withdrawal of clinical drugs. A number of ADRs are caused by undesired interaction of drugs with key proteins involved in normal biological processes. Identification of these ADR-related proteins facilitates the design of drugs with reduced side effects by rationally avoiding unwanted interaction with these proteins. This work explores a statistical-learning-based protein functional classification software SVMDART for identification of potential ADR-related proteins. SVMDART is trained and tested by using 759 ADR-related proteins of different species and 2280 non-ADR-related proteins, testing results show that 93.9% of the ADR-related proteins and 98.2% of non-ADR-related proteins are correctly classified. This suggests that SVMDART may be potentially useful for facilitating the identification of ADR-related proteins.

Submit protein primary sequence for ADR-related protein prediction

Preliminary: Use of "SVMDART" for commercial purposes is not allowed.


Last update: Sep/26/2003

visits