Kai Cao

StriMap models TCR-peptide-HLA interactions using structural and sequence features, achieving state-of-the-art performance and identifying disease-associated peptides in ankylosing spondylitis.

CPS maps physical coordinates to high-fidelity spatial transcriptomics using a multi-scale context distillation strategy, achieving state-of-the-art imputation and denoising performance.

The paper introduces SPICE, a framework that uses splicing assays and generative design to create cell-specific gene regulation elements for research and therapy.

scCausalVI disentangles intrinsic cellular states from treatment effects using a causality-aware generative model, outperforming existing methods in causal inference and multi-source data integration.

Glimmer improves spatial representation learning by adaptively learning graphs that balance molecular similarity and spatial proximity across scales, outperforming existing methods in clustering and spatial conservation.

The paper introduces SONATA, a diagnostic method to detect artificial integrations in diagonal multimodal single-cell data integration, showing its effectiveness across simulated and real datasets.

Graspot integrates spatial transcriptomics data using a graph attention network with optimal transport, improving alignment across datasets and revealing tumor microenvironment structures in breast cancer.

uniPort integrates single-cell multi-omics datasets using a coupled-VAE and Minibatch-UOT, enabling shared latent space embedding and reference atlas construction.

Pamona integrates heterogeneous single-cell multi-omics datasets by aligning cellular structures in a shared space while preserving dataset-specific features.

UnionCom integrates single-cell multi-omics datasets without cell or feature correspondence by aligning topological structures through matrix optimization and global scaling.