deep learning

Thalamic stereo-EEG detects sleep stages N2 and N3 with discrimination approaching scalp EEG

This study shows thalamic stereo-EEG can detect sleep stages N2 and N3 with high accuracy, comparable to scalp EEG, using machine learning classifiers.

Peptorama: Structure creation and exploratory data analysis for non-ribosomal peptides

Peptorama provides tools for creating and analyzing non-ribosomal peptide structures, enabling efficient comparison of known and predicted peptides to aid in biodiscovery.

Rho: A Purpose-Driven Workspace for Human-AI Collaborative Analysis

Rho is a collaborative analysis environment that organizes materials by scientific purpose to support human-AI judgment in bioinformatics.

CARTnet: Template-guided B-cell Epitope Prediction with Contact-aware Attention

CARTnet improves B-cell epitope prediction using a contact-aware deep learning framework, outperforming existing methods on internal and external datasets.

Squalene as a Candidate Signalling Molecule: Computational Evidence for a Predicted KEAP1 Interaction and Implications for Stress-Induced Altruistic Stem-Cell Niche Protection

This paper investigates the structural feasibility of squalene interacting with the KEAP1 Kelch domain using molecular docking and provides a hypothesis for its role in NRF2 regulation under stress.

Generalizable virtual staining from histopathology to spatial proteomics across biological scales

This study introduces QUEST, a model that uses H&E histology to predict spatial protein expression, enabling virtual staining and analysis across biological scales.

ALIGN: A modular framework for quantitative analysis of longitudinal development in Caenorhabditis elegans

ALIGN provides a modular framework for analyzing longitudinal development in C. elegans, enabling accurate segmentation, quality control, and motion correction across imaging conditions.

HiFiTaxa: a reproducible analysis pipeline for full length 16S rRNA gene and ITS amplicon sequencing from PacBio reads

HiFiTaxa is a Nextflow pipeline for analyzing PacBio HiFi reads of 16S rRNA and ITS amplicons, offering multiple classifiers and benchmarking their performance on mock communities and holdouts.

Characterising cortical perturbation-recovery dynamics using TMSEEG: The Brain Dynamic Metric Toolbox

This study uses TMS-EEG and a new toolbox to analyze cortical recovery dynamics after perturbation, finding that extended post-stimulation analysis reveals LICI-related differences in resilience and temporal memory.

Evolution of Vesicle Shape Parameters Based on A Deep-Learning Based Image Analysis Framework Over Extended Time Periods

The paper uses a deep-learning framework to analyze shape dynamics of GUVs over time, measuring morphometric changes and budding behavior.

Disentangling Context-Specific Alzheimer's Disease Progression Signatures through an Explainable AI Framework

ADVISE identifies context-specific Alzheimer's disease progression signatures using an explainable AI framework and validates ROBO2 as a candidate regulator in excitatory neurons.

BioTrajX: evaluating biological plausibility in single-cell trajectory inference

BioTrajX evaluates biological plausibility of single-cell trajectories using three metrics to compare different trajectory inference results.

NeuroWeave-3D: Uncertainty-Aware Evidence Completion for Brain Tumor Segmentation with Missing MRI Modalities

NeuroWeave-3D improves brain tumor segmentation by handling missing MRI modalities through uncertainty-aware evidence completion and consistency checks.

Large-scale AI-guided discovery of antifreeze peptides reveals functional tandem-repeat architectures

This study uses AI to discover antifreeze peptides, finding that tandem-repeat sequences are functional and effective in inhibiting ice recrystallization.

Omics-scale decoding of protein N-glycosylation using protein language model embeddings

The paper decodes protein N-glycosylation sites from protein language model embeddings using a zero-shot framework, revealing biological insights and disease associations.

CloneVI: A deep generative model for deciphering the relationships between cell state and mitochondrial genotype

CloneVI infers clonal structure from cell states in single-cell RNA-seq and ATAC-seq data, linking mitochondrial DNA mutations to specific cell states.

CryoID2: deep learning-enhanced identification of protein and RNA from cryoEM maps

CryoID2 uses deep learning and sequence analysis to identify proteins and RNA from cryoEM maps, achieving high accuracy in molecular identification.

Topographic reference modelling of white matter hyperintensities across the adult lifespan

This paper introduces vTRM, a framework to characterize WMH spatial patterns across age and sex, revealing distinct associations with aging, vascular, and neurodegenerative processes.

A compact motif vocabulary is differentially interpreted across human cell types

The paper shows that a compact set of regulatory motifs can explain gene expression across diverse human cell types through context-dependent interpretation.

Structure-Based Discovery of GTP-Enhanced Rab10 Modulators and Their Effects on Alzheimer Disease-Relevant Cellular Phenotypes

The paper identifies Rab10 modulators through virtual screening and evaluates their effects on Alzheimer's disease-related cellular phenotypes.

Molecular basis of Cyclin B1-CDK1 recruitment by MAD1

The paper identifies a non-canonical helical binding motif in MAD1 that specifically interacts with Cyclin B1 to recruit CDK1 to the SAC, revealing a mechanism for B-type cyclin recognition.

FoXSight: Facilitating Computational Hit Identification in Crystallographic Fragment Screening Experiments

FoXSight identifies weak binding events in crystallographic fragment screening by detecting target-specific density features in noisy electron density maps.

Synthetic reaction-diffusion patterning in mammalian cells

This study designs synthetic reaction-diffusion circuits in mammalian cells to generate spatial patterns with controllable spatial scales.

NPlasmid-seq: Cas9-encoded terminal signatures enable molecule-level nanopore quality control of vector and IVT-template plasmids

NPlasmid-seq detects sub-percent contamination and sequence heterogeneity in plasmid preparations using Cas9-encoded terminal signatures, revealing structural variants missed by consensus-based methods.

Mapping and Probing the Substrate Scope of a Ketoamide-Forming Splicease by Computational Design

The paper uses computational methods to design and evaluate splicease variants for generating diverse ketoamide substrates, finding that top-scoring peptides are spliced more efficiently than single-point mutants.

Extracellular DDR1 targeting remodels collagen-rich tumor stroma and local immunity

This study develops DDR1-targeting nanobodies that remodel collagen-rich tumor stroma and immune compartments, enhancing tumor control without direct tumor-cell inhibition.

Resonant Recognition Model analysis of cGAS: conserved spectral frequencies and a divergent naked mole-rat phase signature

This paper uses the Resonant Recognition Model to analyze cGAS and FANCI sequences across species, identifying conserved spectral features and divergent phase signatures in naked mole-rats.

A validation approach for RNA structures based on deep-learning 3D-structure predictions

This paper extends a protein structure validation method to RNA, showing it can detect errors in low-resolution structures where map-based methods fail.

Optimizing non-invasive assessment of gastric activity: a fully automated channel-wise pipeline for multi-electrode electrogastrography

This paper develops a fully automated pipeline for EGG signal processing to improve artifact rejection and signal quality, showing improved performance over existing methods.

RNA accessibility sequencing reveals transcriptome-wide RBP binding and RNA structure

This paper introduces RAD-seq, a method to quantify RNA accessibility by protein binding, and shows it can detect RBP footprints and RNA structure changes under stress.