Fuhai Li

This paper introduces GiG, a graph AI framework that integrates and interprets whole-person medical and omics data to prioritize biomarkers and improve disease diagnosis.

CellTosg2Sequence integrates biomedical text and signaling graphs into a language model for single-cell analysis, improving performance over existing models.

This paper introduces an autonomous AI agent that performs scientific research tasks end-to-end, demonstrating its ability to generate research artifacts and expose pipeline dynamics.

BioMedGraphica integrates biomedical knowledge into a unified graph and introduces a Textual-Numeric Graph structure for multi-omics analysis.

This paper introduces a framework combining large language models with omics data analysis to prioritize disease targets and drug strategies, validated in Alzheimer's and pancreatic cancer.

SnakeAltPromoter provides a reproducible framework for differential alternative promoter analysis using RNA-seq, benchmarked against CAGE data and scalable for large datasets.

OmniCellAgent autonomously retrieves and analyzes scRNA-seq datasets to generate structured reports and hypotheses across disease settings.

M3NetFlow is a multi-scale multi-hop graph model for multi-omic data analysis that ranks targets and infers signaling pathways, achieving high accuracy in drug response and Alzheimer's disease studies.

This study introduces mosGraphFlow, a graph AI model for analyzing multi-omic data to identify AD-associated biomarkers and signaling pathways, achieving high classification accuracy.

mosGraphGPT is a foundation model for multi-omic signaling graphs trained on cancer data and fine-tuned for Alzheimer's Disease, improving classification accuracy and interpretability.

PathFinder infers multi-cell signaling pathways and communications using a graph transformer model, identifying novel interactions in AD mouse models.

SC2MeNetDrug is a tool that identifies cell-cell signaling targets and predicts drugs using scRNA-seq data to analyze disease microenvironments.

This study identifies four novel genetic loci associated with soluble TREM2 levels in CSF through GWAS and provides functional validation of their roles in Alzheimer's disease.

mosGraphGen generates multi-omics signaling graphs to enable integrative and interpretable graph AI model development by converting multi-omics data into biologically meaningful graphs.

This study uses deep learning models and single-cell RNA-seq data to identify molecular targets and signaling pathways involved in macrophage-fibroblast-cardiomyocyte interactions in heart failure.

This study identifies estrogen's role in suppressing neuro-inflammation and links its loss to increased Alzheimer's disease risk in women through systems pharmacology and signaling network analysis.

This study introduces a novel interpretable k-hop graph attention network to integrate multi-omics data and infer essential targets and signaling networks, showing improved performance over existing models.

The paper introduces DeepSignalingFlow, a deep learning model that predicts drug combination synergy and interprets the mechanisms of synergy using graph neural networks.

The paper develops a deep learning model to analyze the relevance of cancer signaling pathways to patient survival across four cancer types.

The paper introduces a deep learning model to predict signaling cascades by integrating PPI data and omics data, validated using curated KEGG pathways.

This study identifies biomarker genes and potential drugs for G3 medulloblastoma using integrative omics analysis of gene expression, methylation, and copy number data.

This study introduces DeepSignalingSynergy, a simplified deep learning model that predicts tumor cell response to drug combinations using cancer signaling pathways, showing feasibility with NCI ALMANAC data.

This study develops a signaling network model to identify loop sub-structures in PPI networks that predict drug resistance in medulloblastoma subtypes.