deep learning

Brain Network Dynamics from Neuronal Avalanches for Longitudinal Motor-Imagery Decoding

This paper develops a longitudinal BCI framework using neuronal avalanches and graph theory to predict future brain states during motor-imagery tasks, improving decoding accuracy and reducing calibration needs.

LingoCell interprets gene perturbations through natural language and tissue niches

LingoCell interprets gene perturbations using natural language and tissue niches, revealing synergistic disease responses and shared fibroblast programs in cancers.

Accessing Mesoscale Functional Organization from Standard Functional MRI Leveraging AI-Based Super-Resolution

This paper introduces f-LENNS, an AI framework that enhances fMRI resolution without retraining, improving spatial fidelity and generalizing to atypical brain function.

Uncovering Viral Signals Across Cell Populations in Colorectal Cancer

This study maps viral RNA signals across cell types and tissues in colorectal cancer, identifying significant viral associations with specific cell types and tissue conditions.

Dissecting human embryonic patterning during organogenesis reveals a rostral midline progenitor population for mesoendoderm induction

The paper identifies a new rostral midline progenitor population in human embryos using spatial transcriptomics and shows their role in mesoendoderm induction.

KLONE: An Imbalance-Robust k-NN Ensemble for Cross-Modal Single-Cell Clone Assignment

KLONE assigns single-cell RNA-seq cells to DNA-seq identified clones using a k-NN ensemble, achieving high accuracy and minority clone recall in colorectal cancer data.

HV-MultiScaleGNN: A Multi-Scale Graph Neural Network Framework for Host-Virus Protein-Protein Interactions Prediction

This paper introduces HV-MultiScaleGNN, a graph neural network framework for predicting host-virus protein-protein interactions using network topology, achieving strong performance across multiple evaluation metrics.

Comparative larval single-cell transcriptomics provides insights into the evolutionary origins of sensory-neural programmes in Porifera and Cnidaria

The paper compares larval single-cell transcriptomes of sponges and cnidarians to explore the evolutionary origins of neural and sensory cell types.

Time-resolved single-cell screening reveals the dynamic regulatory architecture of human cardiogenesis

This study uses time-resolved single-cell CRISPR screening and a multiome atlas to uncover dynamic gene regulatory networks in human cardiogenesis, identifying stage-specific regulators and CHD-linked gene functions.

OmniT2I: a unified table-to-image framework for survival prediction and cross-modal inference with incomplete multimodal cancer data

OmniT2I converts molecular profiles into images and predicts survival and missing modalities from incomplete cancer data, achieving higher accuracy than separate models.

ASD Neuro Agentic (ANA): An Agentic Framework for Feature Discovery and Cross Site Generalization in Autism fMRI Classification

The paper introduces ANA, an agentic AI framework that improves cross-site generalization in ASD fMRI classification by reducing data leakage and achieving higher accuracy than existing models.

Disentangled generative modeling of single-cell perturbation responses across dynamic cellular states

This paper introduces PertVI, a deep generative model that disentangles treatment effects from cellular state and batch effects in single-cell perturbation data, revealing state-specific transcriptional responses.

Subcellular Spatial Profiling Unveils Links Between Neuritic Plaques with Selective Tau Vulnerability in Alzheimer's Disease

The paper uses subcellular spatial transcriptomics to identify why some neurons in Alzheimer's disease accumulate tau while others resist it, finding differences in molecular programs and neuron-glia interactions.

Ergodic breakdown of brain dynamics under chronnectome complexity across cognitive states and age

The paper introduces a model to classify brain dynamics based on chronnectome complexity, showing high accuracy in distinguishing cognitive states and age-related changes.

Uncertainty-Aware Personalised Gene Expression Prediction Improves Cell-Type-Specific Association Testing

ATTICUS improves gene expression prediction and association testing by incorporating uncertainty awareness and personalized training, enhancing accuracy and cell-type specificity in disease studies.

Accelerating discovery and integration from RNA to Protein with R2T2P

R2T2P integrates transcriptomics and proteomics data to study FSHD, revealing novel RNAs and their impact on translation and protein levels.

Host control and regulation of human anellovirus persistence

This study identifies host factors and cellular contexts governing anellovirus persistence through genomic, transcriptomic, and functional analyses.

MaaD-TAD: an all-atom model of the duplicated PRAMEF/HNRNPCL chromatin domain built from sequence homology

The paper introduces MaaD-TAD, an all-atom model of a duplicated chromatin domain, using sequence homology and cohesin placement to reconstruct DNA structure where Hi-C data are missing.

Benchmarking foundation models for resting-state functional magnetic resonance imaging time series in cognitive aging research

This study benchmarks foundation models for resting-state fMRI time series in cognitive aging, finding mixed results on their utility for cognitive assessment and prediction.

Linking Covariates to State Durations in Dynamic Brain Networks via Hidden Semi-Markov Models

This paper introduces a covariate-dependent hidden semi-Markov model to analyze brain network state durations, finding that age and sex influence dwell times in specific brain states.

Receptor-guided engineering of oncolytic vaccinia creates a synthetic support niche for off-the-shelf NK cell therapy

This paper develops a framework to engineer oncolytic vaccinia virus to support off-the-shelf NK cells by aligning viral payloads with NK cell manufacturing limitations.

Mapping the brain-web morphology to characterise ageing and cognitive impairment

This paper introduces SHARP, a method using cosmological web analysis to characterize brain ageing and cognitive impairment by analyzing MRI data from CamCAN and OASIS-3.

Systematic promoter perturbation reveals limitations of AlphaGenome gene expression predictions

This study evaluates AlphaGenome's ability to predict gene expression changes after promoter perturbations, finding discrepancies between CAGE and RNA-sequencing predictions.

DIALECT: DNA-Programmable Synthetic Cells Communicate with Mammalian Cells through Synthetic Receptors

DIALECT enables synthetic cells to communicate with mammalian cells via DNA-programmed ligand release and engineered receptors, achieving specific signaling and logic gate functions.

Unveiling immune targets via TriStageHLA: high-throughput screening of peptide presentation, peptide-HLA binding, and immunogenicity

TriStageHLA improves epitope prioritization by integrating peptide screening, binding prediction, and immunogenicity assessment with high accuracy and efficiency.

π-SPECFormer: A Mass Spectrometry-based Foundation Model for Decoding the Dark Proteome

This paper introduces π-SPECFormer, a foundation model for mass spectrometry-based proteomics that decodes the dark proteome by pre-training on unannotated spectra and outperforms baselines in peptide sequencing and modification detection.

Synthetic chromatin engineering for STAT1 reactivation toward restoring cisplatin sensitivity in breast cancer

This study reactivates STAT1 in cisplatin-resistant breast cancer using synthetic chromatin engineering, restoring cisplatin sensitivity through enhanced nuclear translocation and transcriptional activation.

Developmentally-regulated splice remodeling shapes functional CACNA1C channel diversity during human neurodevelopment

This study identifies developmentally-regulated alternative splicing of CACNA1C during human neurodevelopment and shows it affects CaV1.2 channel function.

RCDome: A Multi-Database Human Gene Atlas of Regulated Cell Death

RCDome curates a human gene atlas of regulated cell death by integrating multiple databases and validating gene-RCD associations through literature and AI-assisted methods.

scPolyA enables efficient, interactive and agent-guided alternative polyadenylation analysis in single-cell and spatial transcriptomes

scPolyA provides a framework for analyzing alternative polyadenylation in single-cell and spatial transcriptomes, revealing APA patterns in cancer tissues.