Affinity Project - Deep Convolutional Neural Networks That Learn from 3D Images of Proteins
I will give a short overview of the project that other students and I have been working on under the supervision of Dr. Gil Alterovitz at Harvard/MIT Division of Health Sciences and Technology for the last year. I will tell the story behind Afffinity - a virtual screening engine based on deep convolutional neural networks that can predict correct binding positions of drugs with the highest accuracy reported so far.
Maksym Korablyov is a visiting scientist at the Massachusetts Institute of Technology and, jointly, a Fulbright Scholar with Dr. Gil Alterovitz at Harvard Medical School. His work focuses on the development of new methods for drug design, in particular, a high throughput virtual screening engine that utilizes deep convolutional neural networks trained on crystallographic images of proteins in order to make predictions. Maksym holds a MS in bioinformatics from the Georgia Institute of Technology.