Computational Systems Biology and Bioinformatics Lab

We are pursuing multi-faceted projects, either methodology or translationally oriented. These projects are currently supported by grants from the NIH (R01LM011986, R01GM122845, R01AG057555, and R21TR001722).

-         Projects on Methodology Development

  • Deep learning for graph representation and classification
  • 3D deep learning for macromolecular structure analysis
  • Random walk on signed multilayer multiplex network
  • Multi-scale modeling of drug-target binding/unbinding kinetics using Molecular Dynamics simulation and machine learning
  • Machine learning algorithms for predicting genome-scale drug-target interactions and drug phenotypic responses
  • Reconstruction of high-resoulation, proteome-scale drug-target interaction, drug response, and disease-gene association networks through integrating multiple omics data
  • Graphic mining and link prediction of biological networks (in collaboration with Prof. Hanghang Tong at ASU)
  • Text mining of disease and drug response phenotypes (in collaboration with Prof. Zhiyong Lu at NCBI)
  • Differential network and assoication analysis using high-dimensional sparse omics data (in collaboration with Prof. Feng Yang at Columbia Univeristy and Prof. Fuzhong Xue at Shandong University)
  • Compress sensing for signed multilayer network (in collaboration with Prof. Aleksandar Poleksic in UNI)

-         Projects on Translational Sciences

  • Multi-target kinase inhibitor design for anti-cancer therapy with the focus on metastatic prostate cancer and triple-negative breast cancer (in collaboration with Prof. Philip Bourne at UVA, and Prof. Stephen Burley at Rutgers University)
  • Small molecule drug design targeting human microbiome (in collaboration with Prof. Philip Bourne at UVA)
  • Discovery and development of annti-virulence therapies to combat pathogen drug resistance (in collaboration with Prof. Fiona Brinkman at SFU and Prof. David Perlin at Rutgers University)
  • Drug repurposing for Alzheimer's disease by mining multiple GWAS and omics data and using structural systems pharmacology