benchmarking

Efficient approximate Bayesian inference for large, heterogeneous biological datasets

The paper introduces an adaptive minibatch approach for approximate Bayesian inference, improving speed-precision tradeoffs for large, heterogeneous biological datasets.

Quantifying the genetic separability of disease subtypes

This paper introduces GenSep, a framework to quantify genetic separability of disease subtypes, showing current PRSs fall short of theoretical limits.

Sparse Stimulus Generation Improves Reverse Correlation Efficiency and Interpretability

This paper introduces a method for generating sparse stimuli to improve reverse correlation efficiency and interpretability, showing benefits in reconstruction quality, sample size efficiency, and subjective interpretability.

AnnZarro: scalable, interactive exploration of cell-by-cell and gene-by-gene relationships in single-cell data

AnnZarro enables interactive exploration of cell-by-cell and gene-by-gene relationships in single-cell data through a browser-based viewer linked to AnnData objects.

A claim-centred atlas distinguishes biological corroboration from analytical depth in plant stress transcriptomes

The paper reanalyzes plant stress transcriptomes to distinguish biological corroboration from analytical depth using a claim-centred atlas called PLEASE.

InfluProto: Predicting Cross-Species Transmission of Influenza A Virus via Prototype Learning

InfluProto predicts cross-species transmission of influenza A virus using prototype learning, showing high accuracy in host classification and detecting reassortment effects.

UPLiFT: Uncertainty Parameterized Activation for Lightweight Fine-Tuning

UPLiFT improves parameter efficiency in fine-tuning protein language models by optimizing activation space instead of network weights, achieving competitive performance with fewer trainable parameters.

CLARA: Compound Library Active Learning with Reasoning Agents

CLARA uses a language model to make strategy decisions in compound library active learning, outperforming random sampling across multiple protein targets.

Probing isn't Causality in Molecular Models

The paper evaluates whether molecular models can causally represent chemical variables by comparing intervention effects with random controls and pretraining advantages.

Experimental Methods for CRISPR Enzyme Assays with Fluorescence Readout

The paper presents a standardized experimental protocol and analysis framework for fluorescence-based CRISPR enzyme assays to improve reproducibility and consistency in kinetic measurements.

rhythmCI: an R package for rhythm detection using harmonic regression amplitude confidence intervals

rhythmCI provides an R package for detecting circadian rhythms using harmonic regression confidence intervals, improving rhythmicity analysis in small clinical datasets.

simSOMA: a cell-lineage based simulator of the somatic VAF spectrum in plants

simSOMA links plant branching topology to meristem cell-lineage dynamics; in simulations, niche size and turnover drive fixation, while terminal organ growth produces most private low-frequency variants.

Large scale prospective evaluation of co-folding across hundreds of Mac1-ligand complexes and three virtual screens

On 489 evaluable new Mac1 complexes, AlphaFold3, Chai-1 and Boltz-2 placed over half of ligands within 2 Å, but on docking hit-lists AF3 and Boltz-2 added little over docking.

Probing Physical Intuition in Large Language Models through Partner-Dependent Mutation Prediction

This study evaluates LLMs' ability to predict mutation effects based on structural data without using dedicated predictors, finding comparable accuracy to existing methods.

Assessing the potential ecological functions, ecosystem services and disservices of megaherbivore rewilding in traditional rural wood-pastures

This study evaluates the ecological and socioecological impacts of megaherbivore rewilding in traditional wood-pastures, finding context-dependent benefits and risks.

Decision-relevant measurement for precision microbiome interventions

The paper evaluates how well microbiome measurements from different intestinal regions predict intervention decisions compared to stool samples.

Mechanistic prediction of fecal microbiota transplant engraftment and Clostridioides difficile colonization resistance with μODE

This paper introduces muODE, a framework that simulates microbial community dynamics using metagenomics data to predict FMT outcomes and C. difficile resistance.

High-throughput variant libraries and machine learning enable robust target-site selection for toehold-switch design in Detectrons

DeSTAR is a machine learning model that ranks target sites for RNA sensors, improving performance across different transcripts.

AlphaFold3 and Intrinsically Disordered Proteins: Reliable Monomer Prediction, Unpredictable Multimer Performance

AlphaFold3 pLDDT matches AlphaFold2 at calling disordered residues (Matthews correlation 0.693 vs 0.701); on disorder-containing complexes mean DockQ is similar, but interface features explain less variance (42% vs 58%).

Mesodermal-niche interactions direct specification and differentiation of pancreatic endocrine cells in human multilineage organoids

The paper demonstrates that mesoderm-derived cells in organoids promote endocrine commitment and β-cell differentiation in human pancreatic development.

Optimizing Intermittent Theta-Burst Stimulation: The Effects of Pulse Parameters on Corticospinal Excitability

This study tests different pulse parameters in iTBS to determine their effect on corticospinal excitability, finding that long bidirectional pulses are more effective than unidirectional ones.

Robust and replicable effects of ageing on resting state brain electrophysiology measured with MEG.

This paper estimates the full-frequency whole-head profile of age effects on resting-state MEG and shows the effect is replicable and robust across datasets and covariates.

MtbAIM: AI/ML-guided endpoint-specific prediction of small-molecule activity across drug-susceptible and drug-resistant Mycobacterium tuberculosis

Tree models on RDKit/Morgan features predict M.tb activity and MIC endpoints for drug-susceptible and resistant isolates, slightly beating pretrained embeddings; released as a Colab tool.

From Intrinsic Topography to Cognitive Prediction: Benchmarking Approaches to Personalized Cortical Parcellation

The paper benchmarks personalized cortical parcellation methods using fMRI data, finding trade-offs in spatial correspondence, topographic variability, and predictive utility for cognitive traits.

Tensorizing Pairwise Structural Analysis: GPU-Accelerated RMSD and FCC Clustering in HADDOCK3 Without Structural Regression.

The paper accelerates structural analysis in HADDOCK3 by reformulating RMSD and FCC clustering as GPU-based linear algebra operations, achieving significant speedups without compromising accuracy.

Probabilistic Modelling of Base Editing Efficiency and Bystander Outcome Composition

The paper introduces crispAIBE, a probabilistic framework for predicting base editing efficiency and bystander allele composition, achieving high correlation and coverage on benchmark datasets.

Engineered regulatory states resolve context-dependent genetics in macular degeneration

The paper uses CRISPR activation to engineer regulatory states in RPE-1 cells, revealing hidden genetic effects in AMD by analyzing chromatin accessibility and transcriptional interactions.

KEDROS: A Comparative Genomics and High-Resolution Phylogenetic Profiling Workspace

KEDROS is a web platform that performs phylogenetic profiling across proteomes to identify lineage-specific genes and functional divergence, offering a user-defined workflow for comparative genomics.

Transcriptional perturbation effects remain structured beyond the leading global trend

The paper shows that transcriptional perturbations have structured effects beyond a global trend, with specific patterns transferable across datasets.

Linear NeuroPaint: An Efficient Baseline for Cross-Session Neural Alignment and Inpainting

Linear NeuroPaint evaluates a linear model for cross-session neural alignment and inpainting, finding it less accurate than nonlinear models but computationally efficient.