Qi Wang

TopoFuseNet improves drug synergy prediction by integrating hierarchical molecular structures and multi-scale topological features, outperforming existing methods on benchmark datasets.

TransBrain translates brain-wide phenotypes between humans and mice, improving homology mapping and enabling cross-species analysis of brain organization and function.

PI-FC uses state-invariant contrastive learning to extract stable individual brain signatures from fMRI, improving phenotypic prediction accuracy and reducing scanning time across multiple datasets.

This paper introduces STIM, a framework for analyzing brain dynamics during naturalistic stimuli, revealing individual differences in neural landscapes and their links to cognition and mental health.

This paper improves fMRI-based intelligence prediction by using semi-blind machine learning with educational level as supplementary data, achieving better accuracy even with small sample sizes.

This study develops a brain-machine interface for virtual self-motion by decoding neural activity from monkeys performing non-reaching arm movements, showing that different movement classes require distinct decoding strategies.

This paper introduces a new global feature score (GFS) to identify brain regions related to Alzheimer's disease using network analysis, showing it outperforms local features and correlates with cognitive assessments.

HRWR predicts efficacious drug combinations using a hypergraph random walk model, outperforming existing methods in cross-validation and showing promise in cancer treatment.

RWRDC predicts efficacious drug combinations in cancer using random walk with restart, achieving higher AUROC scores than existing methods in breast, colorectal, and lung cancers.

This paper formalizes inter-subject pattern analysis as a multi-source transductive transfer learning problem and demonstrates its benefits through brain decoding experiments.