benchmarking

Calibrating developmental dynamics from single-cell atlases: lineage paths outrun least-action routes

The paper introduces LADD to measure lineage path deviations from inferred developmental dynamics in single-cell RNA-seq data and finds lineages move faster than optimal paths.

Low-fidelity simulator selection for multi-fidelity inference of epithelial vertex models

The paper introduces PRISM to select low-fidelity simulators for multi-fidelity inference in vertex models, finding that early stopping is unreliable while coarser timesteps are effective.

From Fill to Freeze: Redesigning homeRNA to Enable Robust Sampling and Scalable Downstream Processing Workflows for Remote Blood Transcriptomics

This paper validates the redesigned homeRNA platform's robustness to remote study stresses, showing it maintains RNA quality under varied conditions and storage times.

Contrastive Learning Improves Generalization on M2OR Olfactory Binding Dataset

The paper evaluates contrastive learning on VOC-OR binding datasets, showing improved generalization and reduced performance inflation from identity leakage.

Benchmarking Statistical Methods for Differential Composition Analysis in Population-Scale Single-Cell Data

This study benchmarks 12 methods for differential composition analysis in scRNA-seq data, finding LOCOM performs best in controlling false discovery rate and maintaining power.

When perfect is a warning: explanations expose shortcuts that accuracy hides in cancer gene-expression classifiers

The paper shows cancer classifiers can achieve perfect accuracy by learning data source shortcuts rather than biological features, and introduces a method to detect and mitigate this issue.

NTAD: A longitudinal neuroimaging dataset of Alzheimer's disease pathology

The paper introduces a longitudinal neuroimaging dataset of Alzheimer's disease pathology, including multimodal data from patients and controls, to identify and validate novel biomarkers.

The Reasoning Paradox in Scientific Sandboxes: Why Extended Reasoning Fails in Autonomous Bioinformatics and How to Fix It.

This paper identifies two failure modes in extended reasoning models within bioinformatics tasks and introduces DTDR to improve accuracy and reduce errors.

Phylogeny-Guided Global-Local Shrinkage for Microbiome Feature Selection

This paper introduces PGL, a Bayesian regression method for microbiome feature selection that uses phylogenetic trees to improve detection of exposure-associated ASVs.

DECAPPING PROTEIN 5 regulates the circadian clock in Arabidopsis

This study identifies DCP5 as a key regulator of circadian RNA stability in Arabidopsis by modulating the decay of high-amplitude clock-controlled genes.

SpaceBlocks: an expert-supervised workflow for cohort-level spatial transcriptomics analysis

SpaceBlocks is a workflow for reproducible, expert-supervised spatial transcriptomics analysis across platforms, identifying new cell types and differentially expressed genes in multiple datasets.

Top2All: Benchmarking sparse-to-transcriptome reconstruction from high-expression profiles across sex- and sex-steroid-defined conditions

The paper benchmarks sparse-to-transcriptome reconstruction models using high-expression profiles from mouse adrenal cells under sex and sex-steroid conditions, finding that gene-wise metrics and strong baselines are critical for evaluation.

Individual-specific recovery of the healthy human gut microbiota after antibiotics

The paper shows that the human gut microbiota recovers to an individual-specific state after antibiotics, not to a generic healthy state.

Unravelling genomic and environmental contributions to nitrogen-use efficiency in tropical maize through Bayesian causal forests

This paper uses Bayesian causal forests to estimate the causal effect of nitrogen levels on maize traits, finding that environmental factors largely drive trait variation while genomic background influences baseline levels.

Interpreting graph neural networks reveals microbial association patterns in inflammatory bowel diseases

The paper uses graph neural networks to interpret microbial association patterns in IBD, showing that coordinated associations, not just abundance shifts, are predictive.

Reconstructing human gene regulatory evolution using deep learning

This study uses deep learning to predict cis-regulatory effects of human-specific mutations across 305 tissues, revealing their role in gene expression evolution.

SMFnet reveals principles of combinatorial transcription factor assembly

The paper introduces SMFNet and PRISM to analyze how transcription factor assemblies influence chromatin accessibility and identifies sequence-encoded principles of combinatorial TF assembly.

Measurement-Dependent Parameter Recoverability in Neural Cardiovascular LPN Inference

The paper evaluates parameter recoverability in cardiovascular LPN inference using a probabilistic neural framework, showing that accurate hemodynamic prediction does not guarantee correct parameter recovery.

Computational alternative to cardiac biopsy in multiple myeloma: utilizing the shape of routine laboratory data for the classification of AL cardiac amyloidosis

The paper develops machine learning models to classify AL cardiac amyloidosis in multiple myeloma patients using electronic medical records instead of cardiac biopsy.

Transformer Language Models Reveal Distinct Patterns in Aphasia Subtypes and Recovery Trajectories

This paper uses GPT-2 to analyze aphasia subtypes and recovery by examining activation patterns in narrative speech over time, finding distinct differences across subtypes and changes during recovery.

Aneuploidy profiling among different types of brain metastases reveals a colon cancer-specific chromosomal signature

This study identifies a unique aneuploidy profile in colorectal cancer brain metastases compared to other cancer types and metastatic sites.

Evolutionary conservation and divergence of regional fructose-responsive gene regulation in the intestine

This study examines how fructose-responsive gene regulation varies across intestinal regions in mice and fruit flies, finding conserved pathways and divergent metabolic responses.

Tau-CTFP: A connectome-informed continuous-time first-passage model for forecasting the onset of regional tau PET abnormality

This paper introduces Tau-CTFP, a model for forecasting regional tau PET abnormality onset in Alzheimer's disease using connectome-informed continuous-time first-passage modeling, achieving high precision and recall in predicting abnormality onset.

Coordinated Drift of Image Representations in Human EEG

The paper investigates drift in human EEG responses to images across sessions, finding coordinated drift along image identity dimensions and preserved global structure.

Out-of-the-box bioinformatics capabilities of large language models (LLMs)

This study evaluates the performance of three large language models on basic bioinformatics tasks compared to human performance, finding LLMs match human accuracy in some areas but struggle with complex tasks like sequence alignment.

Transcriptomic similarity guides source selection beyond tissue lineage

The paper shows that transcriptomic similarity can rank candidate sources for response prediction better than random, but does not resolve aggregation or transfer uncertainty.

Histone modifications define distinct nucleosome positions in cell-free DNA

This paper identifies histone modification-associated nucleosome positions in cfDNA that correlate with tumor fraction and cancer state across multiple cancers.

Circulating hematopoietic stem and progenitor cells in myeloproliferative neoplasms

This study profiles circulating hematopoietic stem and progenitor cells in MPN patients, identifying transcriptional signatures and quantitative readouts to distinguish MF from other MPNs and predict progression.

Predictive consequences of alternative representations of metabolic measurements: A controlled synthetic study of categories, longitudinal information and transition preservation

The paper evaluates predictive performance of different metabolic measurement representations in synthetic data, finding that some representations outperform others in specific scenarios.

Reference normalizations of Effort-To-Compress across length, alphabet size, and composition

The paper introduces normalized versions of Effort-To-Compress (ETC) to enable fair comparisons of symbolic complexity across different sequence lengths, alphabet sizes, and compositions.