I use and develop computational methods to make sense of large amounts of biological data. Recently, we have developed effective methods for inferring evolutionary couplings in proteins from multiple sequence alignments. We have fed these data into machine learning model to successfully improve the predictive power. We presented the work on using Convolutional Neural Networks as a talk in NIPS 2016. We have applied analogous approach to studying bacterial epistasis in S. pneumoniae, revealing previously unknown insights into genomic basis for penicillin resistance evolution. To learn more about these and other projects, check my Google Scholar profile or drop me an email.