Bo Li

Cancer cells hijack mitochondria from Bhlhe40+ T cells for immune evasion

The paper identifies the Bhlhe40-S100a4 cascade as a key regulator of cancer cell mitochondrial transfer from T cells, offering a target for immunotherapy.

Antigen-aware TCR embeddings reveal conserved immune patterns for personalized immune monitoring

TCRmeta integrates antigen specificity with TCR repertoires to enable cross-individual and longitudinal immune monitoring by revealing conserved patterns and technical variation robustness.

Longitudinal Metaproteomic and Lysine Crotonylation Remodeling Defines Clinical Response to Fecal Microbiota Transplantation in Ulcerative Colitis

Harnessing historical collections to reconstruct the invasion history of Microstegium vimineum (Poaceae)

This study uses historical and contemporary genetic data to reconstruct the invasion history of Microstegium vimineum, identifying multiple independent introductions and genetic adaptations.

A Systematic Comparison of Single-Cell Perturbation Response Prediction Models

The paper benchmarks 13 models for predicting single-cell perturbation responses across 25 datasets, finding performance varies with perturbation effect size and evaluation metric.

An interoperable research agent network for scientific discovery

This paper introduces GUIA, an interoperable research-agent network that enables collaboration between scientific agents across domains, demonstrating its utility in therapeutic target discovery, drug discovery, and spatial proteomics analysis.

A Generalized Supervised Contrastive Learning Framework for Integrative Multi-omics Prediction Models

This paper introduces a supervised contrastive learning framework for multi-omics prediction models, showing improved accuracy over single-omics approaches in type 2 diabetes and high-fat diet studies.

Coronary Artery Disease Transcriptomics Reveals Two Drivers of the Endothelial Cell SR-BI Expression and LDL Transport that Underlie Atherosclerosis

This study identifies two factors, hypercholesterolemia and disturbed blood flow, that upregulate endothelial SR-BI to drive LDL transport and atherosclerosis.

Accessible spatial host-microbe profiling in tumour tissues

This paper introduces HOMES-seq, a spatial transcriptomics workflow for detecting microbes and host transcripts in colorectal tumours, improving sensitivity and reducing costs.

AI-enabled virtual immunopeptidomics links quantitative neoantigen presentation to immunogenicity

The paper develops epiVIP, a deep learning framework to predict HLA-I peptide abundance from (sc)RNA-seq data, showing that peptide abundance and non-selfness jointly determine antigenicity and predict tumor reactivity and survival.

Microglial pruning of extinction-ensemble synapses preserves fear memory

The paper shows microglia preserve fear memory by selectively pruning extinction-ensemble synapses through local signaling cues during memory retrieval.

Spatial Decoding of Tertiary Lymphoid Structure Maturation in Non-Small Cell Lung Cancer Using Deep Neural Networks

The paper develops a framework to decode TLS maturation in NSCLC using spatial omics data and deep learning, identifying two distinct spatial ecosystems linked to patient outcomes.

Aboveground fungal lethal effects of local plants on invasive plant Ageratina adenophora seedlings

This study shows that local plants' aboveground fungi, particularly from phylogenetically close species, reduce germination and survival of invasive Ageratina adenophora seedlings.

KISS1 is a potential immunological and prognostic biomarker for cancer metastasis in pan-cancer and affects breast cancer in a different way

This study investigates KISS1's role in cancer metastasis across multiple cancers, finding it suppresses metastasis in some cancers but promotes it in breast cancer, with implications for immunotherapy and prognosis.

Prober: A general toolkit for analyzing sequencing-based ‘toeprinting’ assays

PROBer is a statistical model that addresses computational challenges in sequencing-based toeprinting assays by learning isoform-specific chemical modification profiles.

Ultrasensitive detection of TCR hypervariable region in solid-tissue RNA-seq data

TRUST detects TCR CDR3 regions in solid-tissue RNA-seq data with high sensitivity, even at low sequencing depth, and is applicable to large tumor cohorts.

Parvalbumin interneuron dysfunction in a thalamus - prefrontal cortex circuit in Disc1 deficiency mice

This study shows that Disc1 deficiency in mice reduces inhibitory drive in the medial prefrontal cortex, linked to impaired feedforward inhibition in the MD-mPFC circuit.

Evaluation of de novo transcriptome assemblies from RNA-Seq data

The paper introduces RSEM-EVAL, a model-based score to evaluate de novo RNA-Seq assemblies, and demonstrates its effectiveness using axolotl limb regeneration data.