Author Description

With the advent of large scale genomic sequencing methods, the precise nucleotide sequence comprising the chromosomes of many organisms has been ascertained. This advance makes the demand of automated genome annotation systems more urgent than ever. Translational initiation sites (TISs) are a type of important gene signals that are critical to the success of annotating genomes. In this book, I will present a novel approach that uses a general multi- agent architecture for accurate TIS prediction. A variety of agents have been employed which facilitate the system to arrive at the solution. A diversified set of techniques ranging from machine learning, evolutionary computation to swarm intelligence have been explored in our implementation. The experimental results on nine different data sets have demonstrated the advantage of using the proposed approach. A recommendation system that helps identify the optimal parameter settings for the system has been introduced which helps increase the efficiency of the entire model.