COMPARISON OF COMPUTATIONAL MODELS IN THE PREDICTION OF ADVERSE DRUG REACTIONS

Abstract

Computational pharmacology is the application of bioinformatics and computational biology with relevance to pharmacology, including understanding of drug action, adverse drug reaction, identification of drug targets, and drug design. Early and accurate identification of adverse drug reactions (ADR) is critically important for drug development and clinical safety. Often, the adverse effects of drugs are not discovered until years later, after the drugs’ release to the market. The post hoc analysis is usually unable to detect rare or delayed onset ADR until clinical evidence accumulates. The process of drug development and ADRs discovery takes years, meaning that a lot of harm would have been caused to lives before evidences are accumulated, therefore, developing a computational pharmacology model that can be used to make informed decisions to reduce the rate of attrition in drugs under development and increase the number of drugs with an acceptable benefit/risk ratio is paramount. This paper reviews the computational methods of the few works that have been used so far to address this issue and hence looks at the applicable methods that can be used in drug to ADR prediction.