Machine Learning for Intelligent Bioinformatics - Part 1 Machine Learning Integration

The highly-interdisciplinary field of bioinformatics goal is to develop computing systems capable of analyzing molecular biology. We argue that bioinformatics has undergone a historical transition from the first phase to the second, now underway. The first phase was dominated by the use of traditional, intelligence-free computer programs such as database management systems, on the one hand, and by a small fraction of computational statistics, on the other hand. The second phase, now unfolding, heavily relies on artificial intelligence techniques such as probabilistic, nearest neighbor and genetic algorithm approaches, inter alia. In this first part of the present work, we describe both phases, emphasizing integration of alternative machine learning methods such as grammatical inference. This helps in constructing an overall framework including intelligent control described in the second part of the work, reported in an independent paper.