Collaborative Matrix Factorization for Predicting Drug-Target Interactions

Title: Collaborative Matrix Factorization for Predicting Drug-Target
Interactions

Abstract: Computationally predicting drug-target interactions is
 useful to discover potential new drugs (or targets). Currently,
 powerful machine learning approaches for this issue use not only
 known drug-target interactions but also drug and target
 similarities. Using similarities is well-accepted pharmacologically,
 since the two types of similarities correspond to two recently
 advocated concepts, so-called, the chemical space and the genomic
 space. In this talk, I will first briefly review the literature of
 similarity-based machine learning methods for this issue and then
 present our recent method, which can overcome problems in existing
 methods. Our method is based on a factor model, named Multiple
 Similarities Collaborative Matrix Factorization (MSCMF), which has
 the following two key ideas: 1) MSCMF allows to incorporate more than
 one similarity matrices over drugs as well as those over targets, where
 weights over the multiple similarity matrices are estimated from data
 to automatically select the similarities, which are effective for
 performance improvement. 2) MSCMF projects drugs and targets into a
 common low-rank feature space (matrix), which is estimated to be
 consistent with weighted similarity matrices over drugs and those
 over targets by an alternating least squares algorithm. I will finally
 explain experimental results, obtained by using both synthetic and
 real datasets, by which the performance of MSCMF was extensively
 evaluated.
We note that our approach is general and applicable to any binary
 relations with similarities over elements, which can be found in many
 applications, such as recommender systems. In fact, MSCMF can be
 regarded as an extension/generalization of weighted low-rank
 approximation for one-class collaborative filtering.

References:
1. Ding, H., Takigawa, I., Mamitsuka, H. and Zhu, S., Similarity-based
Machine Learning Methods for Predicting Drug-target Interactions: A
Brief Review. To appear in Briefings in Bioinformatics.
2. Zheng, X., Ding, H., Mamitsuka, H. and Zhu, S., Collaborative
Matrix Factorization with Multiple Similarities for Predicting
Drug-Target Interactions. Proceedings of the Nineteenth ACM SIGKDD
International Conference on on Knowledge Discovery and Data Mining
(KDD 2013)., pp. 1025-1033, Chicago, IL, USA, Aug. 2013, ACM Press.


About the speaker:
Hiroshi Mamitsuka received the B.S. degree in biophysics and biochemistry, the M.E. degree in information engineering, and the Ph.D. degree in information sciences from the University of Tokyo, Tokyo, Japan, in 1988, 1991, and 1999, respectively. He is involved in research on machine learning, data mining, and bioinformatics. His current research interests include mining from graphs and networks in biology and chemistry.

http://www.bic.kyoto-u.ac.jp/proteome/mami/