NCSA Fellow Develops New Deep Learning Tool

The DeepContact tool uses neural network techniques to predict protein structures

Jian Peng, National Center for Supercomputing Applications (NCSA) fellow and Assistant Professor of Computer Science at the University of Illinois, and U of I computer science graduate student Yang Liu have discovered a new technique for using deep learning to predict the structure of proteins.

According to a report at HPC Wire, the research led to the development of the DeepContact deep learning tool, which “… automatically leverages local information and multiple features to discover patterns in contact map space and embeds this knowledge within the neural network.”

Misfolding proteins can lead to a number of diseases, including Alzheimer’s. Peng and Liu will use DeepContact to improve models for predicting protein folding.