Hanchen Wang

Archetype analysis of multiplexed antibody imaging reveals interpretable morphological programs in optical pooled screens

The paper introduces CellMAPP and CHART to analyze multiplexed antibody imaging data, revealing modular morphological programs in response to genetic perturbations.

An Integrated Atlas of the Human Kidney Spanning Health and Diseases

The paper integrates multiple single-cell and single-nucleus RNA-sequencing datasets to create a comprehensive Human Kidney Cell Atlas, identifying cell types, states, and disease-associated populations.

AutoScreen: AI Co-Scientist System for Target Discovery in Functional Genomics

AutoScreen improves gene prioritization in functional genomics by integrating AI agents for pre-screen design and post-screen analysis, achieving higher hit recovery and precision than existing methods.

Perturb-ME: Scalable mechanism discovery from phenotype-enriched genome-wide screens

Perturb-ME combines CRISPR screening and single-cell profiling to identify regulatory modules governing gene programs in melanoma.

Orion: Towards Lab Automation with Computer-Using Agents

Orion, an AI agent, automates biomedical image analysis by integrating large language models with software execution and reasoning, generating research reports and hypotheses from imaging data.

Spatiotemporal profiling reveals the role of inflammatory niche in driving prostate cancer

This study uses single-cell and spatial profiling to show that inflammatory signals from macrophages and fibroblasts drive CD8 T cell exhaustion and resistance to immune checkpoint blockade in prostate cancer.

Learning multi-cellular representations of single-cell transcriptomics data enables characterization of patient-level disease states

PaSCient learns patient-level disease representations from scRNA-seq data, enabling disease classification and downstream analyses like dimensionality reduction and patient subgroup discovery.

Metric Mirages in Cell Embeddings

The paper shows that current cell embedding evaluation metrics can be misleading and introduces a new metric, scGraph, to detect distortions in embeddings.

Biomni: A General-Purpose Biomedical AI Agent

An LLM agent with a literature-mined environment of 150 tools, 105 packages and 59 databases plans in code, matching reported human accuracy on database and sequence question answering.

PerTurboAgent: A Self-Planning Agent for Boosting Sequential Perturb-seq Experiments

PerTurboAgent predicts gene panels for sequential Perturb-Seq experiments using an LLM-based agent, outperforming existing strategies in identifying phenotypically impactful genes.

SpatialAgent: An Autonomous AI Agent for Spatial Biology

SpatialAgent integrates large language models with dynamic tool retrieval, multimodal interpretation, and claim verification to automate end-to-end spatial biology workflows from gene-panel design to tissue annotation and hypothesis generation.

Autonomous AI Agents Discover Aging Interventions from Millions of Molecular Profiles

The paper reanalyzes millions of molecular profiles to identify aging interventions using AI agents, finding over 500 interventions that reduce biological age and revealing patterns in aging modulation.