Anjun Ma

This paper introduces mmFEA, a kinetic-aware framework for estimating metabolic flux changes from transcriptomics and metabolomics data, showing improved accuracy over baseline methods.

The paper shows that TREM2-dependent anabolic programs in microglia decline with advancing Alzheimer's disease, linking anabolic capacity to microglial resilience rather than activation state.

This study constructs a cross-model spatial and single-cell atlas of PKD, identifying Spp1 (osteopontin) as a conserved driver of disease progression through its signaling to Ly6clo monocytes.

The paper constructs a multi-omics single-cell atlas of human RPE and choroid, identifying cell types, aging patterns, and genetic risk factors for AMD.

SpaFlow models ligand-receptor interactions in spatial transcriptomics data by simulating diffusion and binding processes, revealing spatially coherent signaling patterns in multiple disease contexts.

This paper introduces BioGuider, a multi-agent platform that evaluates and improves bioinformatics software documentation, showing a link between documentation quality and citation frequency.

This paper develops TDMA, a transcriptomics-based model to estimate DNA methylation activity and epigenetic aging from bulk, single-cell, and spatial transcriptomics data.

This paper introduces scPEFT, a parameter-efficient fine-tuning framework for scLLMs that improves task adaptation with reduced computational costs and enhances performance in disease-specific and cross-species analyses.

This study examines epigenetic, microbiome, and immune differences in early-onset colorectal cancer (EOCRC) compared to average-onset CRC, finding accelerated epigenetic aging and stronger immune-microbe correlations in EOCRC.

The paper identifies PRDM6 as a promoter of tumor progression in HNSCC by modulating immune gene expression using scRNA-seq data and IRIS3 analysis.

TrimNN identifies spatial cell organization patterns from spatial transcriptomics and proteomics data to reveal associations between cell-type arrangements and phenotypes.

This paper introduces DeepSAS, a deep graph framework to identify senescent cells and their associated genes, revealing cell-type-specific senescence features in aging and disease.

scGNN+ integrates ChatGPT with graph neural networks to improve code optimization, tutorial generation, and visualization in single-cell research.

This paper introduces MICAH, a graph neural network framework that identifies cancer-associated microbial communities in intratumoral microbiome data across five cancer types.

STREAM infers enhancer-driven gene regulatory networks from single-cell RNA-seq and ATAC-seq data, showing improved performance in TF recovery and enhancer-gene discovery.

ssREAD is a database that compiles single-cell and spatial RNA-seq data for Alzheimer's Disease, offering analysis tools for integrative studies.

The paper introduces MEGA, a tool to identify cancer-associated microbes using RNA-seq data from the ORIEN dataset, leveraging graph attention networks and microbial community information.

MarsGT identifies rare cell populations in multi-omics single-cell data, outperforming existing tools and revealing novel biological insights in human and mouse datasets.

This paper introduces IDAM, a method for inferring disease-associated microbial gene modules from metagenomic and metatranscriptomic data, showing superior performance and reproducibility in inflammatory bowel disease datasets.

This paper introduces SpaGFT, a graph Fourier transform model for spatial omics data analysis, enabling accurate identification of tissue modules and their functional relationships in spatial transcriptomics and proteomics data.

This paper introduces PHAT, a deep graph learning framework for peptide secondary structure prediction, achieving better accuracy and interpretability for short peptides.

This study shows that aging of CNS-resident cells worsens EAE severity and reduces remission in mice, suggesting microglial activation and aging properties as therapeutic targets for progressive MS.

GraphPred predicts multiple DNA motifs from ATAC-seq data using a graph neural network and coexisting probability, outperforming existing methods on 88 datasets.

RESEPT uses deep learning to visualize and characterize tissue architecture from spatial transcriptomics data, revealing pathological structures in Alzheimer's and glioblastoma samples.

DeepMAPS infers biological networks from single-cell multi-omics data using a heterogeneous graph transformer, outperforming existing tools in cell clustering and pathway prediction.

This study evaluates 20 deep learning models for cis-regulatory motif finding using genomic data, finding that model choice depends on data type and size.

This paper introduces scDEAL, a deep transfer learning framework that predicts cancer drug responses from single-cell RNA-seq data by integrating bulk RNA-seq drug response data and interpretable gene signatures.

scGNN introduces a graph neural network framework for scRNA-Seq analyses, improving gene imputation and cell clustering on benchmark datasets and revealing disease-related neural development in Alzheimer's disease.

This paper introduces a deep forest-based method for predicting protein-protein interactions, achieving high accuracy in yeast and Helicobacter pylori and improving drug discovery.