Multi-Class Protein Fold Prediction Using Stochastic Logic Programs ?

This paper presents an application of stochastic logic pro- grams (SLPs) to learning probabilistic logic rules in protein fold predic- tion. We apply SLP parameter estimation algorithm to a previous study in which rules have been learned by inductive logic programming (ILP). On the basis of experiments, we demonstrate that probabilistic ILP ap- proaches (eg. SLPs) have advantages for solving multi-class protein fold prediction problems and SLPs have outperformed ILP plus majority class predictor in both predictive accuracy and result interpretability.