Discover, rate, and discuss biology preprints.

TDKC improves metagenomic pathogen detection by reducing memory and computational costs while maintaining accuracy.

MaskTalk improves cell-cell communication inference by masking non-target cells, showing better effect sizes and shorter distances in breast cancer spatial transcriptomics data.

This study evaluates spatial transcriptomics foundation models' ability to capture multi-scale spatial variation using SAFFRON, finding that one model better captures global gradients but not local patterns.

The paper introduces COMPASS, a model that decomposes genetic perturbation responses into shared and gene-specific components, showing improved accuracy and discrimination compared to existing methods.

The paper introduces PHORAGER, a pipeline for identifying prophages in bacterial genomes, and reports the discovery of 23,132 putative prophages in ESKAPE pathogens.

BENDER DB compiles predicted protein binding sites from neglected disease pathogens using AlphaFold structures and multiple predictors, offering a consensus and visualization tools to aid drug discovery.

PLMView is an unsupervised framework for fine-grained protein function classification from sequence, achieving high precision and speed without labeled data or model fine-tuning.

This paper explores the use of LLM-based coding agents in scientific computing, presenting case studies in the life sciences and discussing their potential and challenges.

The paper identifies short linear motifs in Toxoplasma gondii secreted proteins using a computational pipeline and validates some predictions experimentally.

LightAlign is a lightweight HiFi read aligner that reduces memory usage for genome assembly in resource-constrained settings.

The paper evaluates how using a sample-matched reference genome affects single-cell transcriptomic analysis in Plasmodium falciparum, finding that var genes require a matched genome for accurate expression inference.

This paper introduces CREAM, a deep learning model for differential expression prediction using personalized genomes, showing improved performance on training genes but challenges in generalizing to unseen genes.

This study introduces BIGraph-ST, a method that integrates pathway activity and histological features to improve spatial domain identification in spatial transcriptomics, showing better performance in cancer tissues.

BirdCODE detects and classifies bird vocalizations from large audio datasets, enabling large-scale studies of bird communication with improved temporal resolution and species coverage.

MEGA-ODE reconstructs dynamic omics landscapes from sparse data, predicts unmeasured states, and prioritizes perturbations toward biological endpoints.

The paper introduces CARS, a transferable force field for carotenoids that improves simulation accuracy across diverse chemical structures and outperforms existing force fields.

Hoike generates condition-specific bulk transcriptomes using a joint-embedding predictive architecture and diffusion models, conditioned on normal tissue profiles.

This paper calibrates epigenetic age between EPICv2 and MSA arrays using linear regression models to reduce cross-platform differences in six epigenetic clocks.

IMAS identifies target-specific regulatory architectures from multiomic tumour data by integrating RNA and chromatin-accessibility profiles to uncover perturbation-sensitive mechanisms.

This paper develops a deep learning framework for obstructive sleep apnea classification from ECG signals, using two datasets and integrating feature engineering with transfer learning for improved generalization and personalization.