Computational Biology and Biophysics Lab

Our mission and vision

We apply and design computational methods that integrate hypothesis- and data- driven approaches, including machine learning/deep learning, molecular modeling and molecular dynamics simulations. We work in close collaboration with experimental groups and develop hybrid integrative approaches that use experimental data - ranging from hydroxyl radical footprinting, chemical crosslinking to cryo-electron microscopy - to guide us in molecular modeling and simulations. This multi-faceted approach can offer experimental leads for identifying driver mutations and genes, and for designing drugs that affect chromatin-related pathways.

Research summary

• Introducing a novel deep learning method for predicting nucleosome positioning
• Generating a prediction model based on large-size data
• Providing a web server and making the algorithm ready to use