Discovering Multi-Label Hierarchical Classification Rules for Protein Function Prediction

This work proposes the use of an Artificial Immune System (AIS) for predicting biological protein functions described in the gene ontology. The main challenge of this work is to discover a set of classification rules that are both hierarchical and multi-label, in the sense that a single classification rule can assign to a protein several classes (labels) in several levels of the gene ontology's hierarchy. The performance of the proposed algorithm is evaluated considering 3 protein data sets.