Discover, rate, and discuss biology preprints.

The paper uses single-sample network analysis to identify pathogenic hubs in Sjögren's disease and prioritize drug targets.

This paper evaluates statistical methods for analyzing repeated-measures proteomics data, finding that unstructured covariance models outperform conventional approaches in handling variance heterogeneity and within-subject correlations.

The paper introduces phylogenetic parallelograms to visually compare multiple discordant phylogenetic trees and applies them to study introgression and organellar discordance.

The paper introduces resolving power as a metric for protein stability prediction, showing it correlates better with model performance than traditional benchmarks.

This paper argues for evaluating protein language models through prospective experimental success rather than retrospective predictive accuracy.

CycDelta is a multimodal residue-graph model for predicting permeability changes in cyclic peptides by comparing within-source permeability differences, outperforming baselines on out-of-distribution and zero-shot tasks.

DeepCDS improves coding sequence prediction in short prokaryotic sequences using deep learning and protein language model embeddings, outperforming existing methods across various sequence lengths and error profiles.

ColBERT-PPI predicts protein partners and their binding sites at the residue level using a multi-vector framework, outperforming existing methods on benchmark datasets.

The paper trains a machine-learning model on native protein structures to predict interaction preferences and energetics without using affinity or mutation labels, achieving performance comparable to existing methods on a benchmark.

The paper introduces Cerebra-Epistasis, a framework that improves multi-mutant fitness prediction by modeling epistatic effects and integrating structure-aware sequence analysis.

Culsma provides an executable specification language for laboratory protocols, enabling both human readability and machine execution to ensure precise experimental execution and verification.

The paper introduces a lossless compression algorithm for protein databases, improving memory efficiency and classification accuracy in Centrifuger for metagenomic analysis.

This paper introduces DualTopoDDI, a deep learning model for predicting drug-drug interactions with mechanistic insights, achieving high accuracy and interpretability.

The paper introduces MeRN, a graph-guided variational autoencoder for learning single-cell metabolic states from transcriptomes, and identifies metabolic drivers in various biological contexts.

FlowMap jointly models cellular states and dynamics in a geometric framework, recovering interpretable trajectories and developmental patterns from single-cell RNA sequencing data.

The paper identifies 20 new protein kinase-like families in the human oral microbiome using structure prediction and sequence analysis, revealing evolutionary strategies like enzymatic specialization and structural co-option.

Cell-Lens improves LLM reasoning on single-cell data by using a structured representation that captures more biological content than gene lists alone.

The paper introduces ATACofthesnake, a scalable framework for analyzing multifactorial and time course chromatin accessibility data, providing a structured workflow to handle such data reproducibly.

The paper introduces CAMP-14-3-3, a framework that improves 14-3-3 binding prediction by integrating multi-scale biological features and protein language model embeddings, outperforming existing methods.

Open-source STAR binary for five transcriptomic assays runs 3.9- to 6.2-fold faster than Cell Ranger 9.0.1 for scRNA-seq and Perturb-seq, 17- to 41-fold for Flex, with closely concordant counts.