Hao Li

AMD1-driven MTA production sustains chemoresistance by rewiring PMN-MDSC immunosuppression

This study identifies AMD1-driven MTA production as a mechanism sustaining chemoresistance by enhancing PMN-MDSC immunosuppression.

ART: A Compositional Physical Theory for the End-to-End Design of Solid-Phase Immunosensors

The paper introduces ART, a physical theory for designing solid-phase immunoassays by modeling coupled processes and demonstrating its application through examples.

LigRNA: High-accuracy prediction of RNA-ligand binding sites via dynamic chemical complementarity and language model encoding

LigRNA predicts RNA-ligand binding sites using a graph neural network that combines language model embeddings and physicochemical features, achieving improved performance over existing methods.

Wetness modulates the effects of grazing on net ecosystem productivity in global grasslands

This study shows light grazing increases net ecosystem productivity with higher wetness, while heavy grazing reduces productivity regardless of wetness in global grasslands.

Therapeutic Potential of Hypoxia-Preconditioned hiPSC-Epicardial Cell-Derived Exosomes in Mice with Myocardial Infarction

This study compares the cardioprotective effects of normoxic and hypoxic exosomes from hiPSC-derived epicardial cells in MI mice, finding that hypoxic exosomes provide superior protection through miR-214-3p-mediated mechanisms.

LadybirdBase: a comprehensive biology, ecology and omics resource for ladybird beetles (Coccinellidae)

LadybirdBase is a database compiling omics data and biological information on ladybird beetles, offering tools for species identification and omics analysis.

A Population-scale Single-cell Spatial Transcriptomic Atlas of the Human Cortex

This study creates a single-cell spatial transcriptomic atlas of the human cortex, revealing gene expression patterns, aging effects, and disease associations across layers and species.

Hierarchical affinity landscape navigation through learning a shared pocket-ligand space

LigUnity is a foundation model that jointly embeds ligands and protein pockets into a shared space to improve affinity prediction in virtual screening and hit-to-lead optimization.

Yeast aging from a dynamic systems perspective: Analysis of single cell trajectories reveals significant interplay between nuclear size scaling, proteasome dynamics, and mitochondrial morphology

The paper analyzes single-cell yeast aging trajectories to show proteasome dynamics and nuclear size scaling are key predictors of lifespan and influence mitochondrial morphology.

ElixirSeeker: A Machine Learning Framework Utilizing Attention-Driven Fusion of Molecular Fingerprints for the Discovery of Anti-Aging Compounds

ElixirSeeker uses attention-driven fusion of molecular fingerprints to identify anti-aging compounds, finding four that extend C. elegans lifespan.

Unlocking the Potential: FKK6 as a Microbial Mimicry Based Therapy for Chronic Inflammation-Associated Colorectal Cancer in a Murine Model

This study evaluates FKK6, a microbial mimic of an indole metabolite, as a therapy for colitis-associated colorectal cancer in mice, showing anti-tumor effects and safety.

An organ-based multi-level model for glucose homeostasis: organ distributions, timing, and impact of blood flow

The paper updates a multi-level mathematical model of glucose homeostasis to better reflect organ-specific glucose uptake, dynamics, and the impact of blood flow on insulin-stimulated glucose uptake.

Rewiring the specificity of extra-cytoplasmic function sigma factors

The paper deciphers the DNA-specificity code of ECF σ factors to predict their regulons and reveals their varied regulatory roles in bacterial gene regulation.

New insights on human essential genes based on integrated analysis

This study integrates multiple analyses to characterize human essential genes and develops a web platform for their comprehensive interpretation.

iFORM: incorporating Find Occurrence of Regulatory Motifs

iFORM is a tool that improves the accuracy and sensitivity of TF motif scanning by integrating results from multiple motif discovery programs.

Genome-wide identification and characterisation of HOT regions in the human genome

The paper identifies and characterizes HOT regions across human cell types, showing their association with developmental genes and unique epigenetic features.

PEDLA: predicting enhancers with a deep learning-based algorithmic framework

PEDLA uses deep learning to predict enhancers from heterogeneous data, achieving high accuracy across multiple cell types.