AI for Biomedical Data

Deep learning, probabilistic models, and generative methods for multi-omics, regulatory genomics, and clinical genomics.

The lab develops AI methods tailored to the structure of biomedical data, where datasets are heterogeneous, noisy, and biologically constrained. We work across deep-learning, probabilistic, and emerging generative approaches, with applications in regulatory genomics, single-cell trajectory inference, long-read variant interpretation, and translational genomics.

Foundational work includes deep-learning imputation for transcription factor binding (Qin & Feng, 2017), integrative regulatory modeling with Lisa (Qin et al., 2020), and statistical integration methods for high-dimensional, multi-source omics data (Zang et al., 2016). We also contribute to community benchmarking of open problems in single-cell analysis (Luecken et al., 2025) and to reviews on inferring developmental trajectory and cell fate (Wang et al., 2021), ensuring our methods are evaluated against the strongest baselines.

References

2025

  1. Nat Biotechnol
    Defining and benchmarking open problems in single-cell analysis
    Malte D Luecken, Scott Gigante, Daniel B Burkhardt, and 14 more authors
    Nature Biotechnology, Jul 2025

2021

  1. Curr Opin Syst Biol
    Current progress and potential opportunities to infer single-cell developmental trajectory and cell fate
    L Wang, Q Zhang, Qian Qin, and 1 more author
    Current Opinion in Systems Biology, 2021

2020

  1. Genome Biol
    Lisa: inferring transcriptional regulators through integrative modeling of public chromatin accessibility and ChIP-seq data
    Qian Qin, J Fan, R Zheng, and 8 more authors
    Genome Biology*Co-first authors , 2020

2017

  1. PLoS Comput Biol
    Imputation for transcription factor binding predictions based on deep learning
    Qian Qin and Jianfeng Feng
    PLoS Computational Biology, 2017

2016

  1. Nat Commun
    High-dimensional genomic data bias correction and data integration using MANCIE
    C Zang, T Wang, K Deng, and 11 more authors
    Nature Communications, 2016