Development and evaluation of machine learning algorithms for biomedical applications

Gene network inference and drug response prediction are two important problems in computational biomedicine. The former helps scientists better understand the functional elements and regulatory circuits of cells. The latter helps a physician gain full understanding of the effective treatment on patients. Both problems have been widely studied, though current solutions are far from perfect. More research is needed to improve the accuracy of existing approaches.This dissertation develops machine learning and data mining algorithms, and applies these algorithms to solve the two important biomedical problems. Specifically, to tackle the gene network inference problem, the dissertation proposes (i) new techniques for selecting topological features suitable for link prediction in gene networks; a graph sparsification method for network sampling; (iii) combined supervised and unsupervised methods to infer gene networks; and (iv) sampling and boosting techniques for reverse engineering gene networks. For drug sensitivity prediction problem, the dissertation presents (i) an instance selection technique and hybrid method for drug sensitivity prediction; (ii) a link prediction approach to drug sensitivity prediction; a noise-filtering method for drug sensitivity prediction; and (iv) transfer learning approaches for enhancing the performance of drug sensitivity prediction. Substantial experiments are conducted to evaluate the effectiveness and efficiency of the proposed algorithms. Experimental results demonstrate the feasibility of the algorithms and their superiority over the existing approaches.