Research Projects

Metabolite Annotation via Mass Spectra Prediction

Develop an ensemble model based on graph neural network (GNN) and multi-layer perceptron (MLP) for muti-label regression. Enhance the model by applying attention and multi-tasking to address label dependency.

Enzyme-substrate Interaction Prediction

Design a general framework for recommender systems via multi-task learning. Develop enhanced neural network-based recommender systems to predict enzyme-substrate interactions.

Antibody Sequence Analysis and Design

Evaluate antibody features through machine learning methods, including random forest, SVM, and AdaBoost. Develop a novel pipeline to design antibody sequences using decision tree.

Drug Concentration Prediction

Develop a neural network model for time series regression on drug concentration. The model used for precise drug analysis to facilitate drug review procedures.