The machine learning (ML) approach aims at designing/programming machines/computers that can learn like a human; that is, learn from experience and discover information from the available data, but are capable of working with much larger data sets and learning at a significantly accelerated rate. The computers are programmed to optimize performance criteria by using example data or past experience. The optimized criterion can be the accuracy provided by a predictive model — in a modelling problem — or the value of a fitness or evaluation function — in an optimization problem.
This approach is applicable to bioinformatics because the subjects are highly complex biological systems. Additionally, molecular biology that depends on a lot on experimental data can be adapted for machine learning approaches. In fact, one of the earliest application areas in machine learning was molecular biology: Stormo and his colleagues used the perceptron algorithm to locate the initial translation sites in E. coli. (Stormo et. al. 1982).
This work is licensed under a Creative Commons Attribution-NonCommercial-ShareAlike 4.0 International License.