Danqing Yin

SpaTranslator translates spatial omics data across modalities using a deep generative framework, outperforming baselines in clustering and biological coherence.

This paper benchmarks three LLMs for extracting gene-disease-phenotype relationships from biomedical texts, finding GPT-4o achieves 88.8% accuracy using zero-shot prompting.

AtlasAgent uses a vision-language model to rapidly evaluate large-scale single-cell data integration, outperforming traditional methods in speed and efficiency.

Vulture is a scalable cloud-based pipeline for microbial calling in scRNA-seq data, outperforming existing tools in speed and cost while discovering virus-enriched cell types in cancer and COVID-19 datasets.

Scope+ provides an open-source architecture for accessing and analyzing large single-cell atlases at the cell level, demonstrated through the Covidscope portal for COVID-19 data.