benchmark study

RepGene: Auditing Condition-Specific Fusion Biases for Multi-View Gene Representations

The paper evaluates multi-view gene representations under a matched contract, showing condition-dependent fusion operator rankings across different view combinations and tasks.

Benchmarking Uncertainty and Improving Risk Ranking in Multi-task Bioactivity Prediction

The paper benchmarks uncertainty estimation and risk ranking in multi-task bioactivity prediction, introducing ICSR to improve reliability of kernel-based predictions.

A fibroblast-rich transcriptomic axis reflects stromal remodeling within the CMS4 phenotype in colorectal cancer

The paper evaluates a 87-gene score in colorectal cancer, finding it correlates with stromal remodeling and CMS4 but shows inconsistent prognostic value across cohorts.

How Similar Are Two Brains? A Comprehensive Benchmark of Brain Network Similarity Measures

This paper benchmarks 16 brain network similarity measures, finding DeltaCon is most accurate for parameter recovery but no measure excels across all criteria.

GO-Term Enrichment of Proteome-Scale Docking Profiles as a Biological Search-Space Reduction Layer for Protein Target Discovery

GO enrichment of proteome-scale docking profiles shrinks the candidate target list: a held-out GO gate kept 11.32% of known targets in 1.26% of candidates; additive reranking was mixed.

Passenger co-deletion confounds glutaminolysis signatures anchored on PTEN loss: a cautionary case for location-aware signature design

The paper shows that a glutaminolysis signature correlates with PTEN copy-number loss not due to metabolic changes but because of passenger co-deletion, highlighting the need for location-aware signature design.

A Comparative Benchmark of Biomedical Language Models for Concept Normalization from Real-World Text

This study benchmarks biomedical language models for concept normalization from real-world text, finding Qwen 2 Instruct improves accuracy when combined with SapBERT.

MAT-classifier: A memory-efficient pipeline for accurate genus level profiling from ancient metagenomic data

Per-genus Bowtie2 or BWA-aln alignments in place of MALT cut alignment-stage memory from about 140 GB (aMeta) to about 7 GB and gave higher simulated-data AUC (0.87 vs 0.79).

Measurement reliability bounds functional benchmarks and relocates where variant effect prediction fails

The paper benchmarks variant effect predictors using MAVE data, identifies reliability ceilings, and shows that correction improves performance in some regions but not others.

Latent generative search unlocks de novo design of untapped biomolecular interactions at scale

Reward-guided inference-time search over the Proteína-Complexa sequence-structure codesign model yields validated binders across 127 targets, kinases and a free carbohydrate, with codesigned sequences outperforming ProteinMPNN redesign.

Incomplete references leave bulk deconvolution targets non-identifiable, but identification has a computable precision price

With an incomplete deconvolution reference, operator history changes estimates and associations, and the target is not point-identified; fitdrop scores remedies on shrinkage, coverage, certification and false certification, and prices measurements.

Behaviorally prioritized entity-relation structure captures human visual cortical representations of natural scenes

The paper uses triplet-based semantic representations to capture structured relational information in human visual cortex during natural scene perception, showing better alignment with human high-level visual areas than macaque inferotemporal cortex.

NM-RARe: A FRAMEWORK FOR REGIONAL ARTHROPOD CONSERVATION ASSESSMENT: INSIGHTS FROM NEW MEXICO

NM-RARe is a regional database for arthropod conservation in New Mexico, compiling species data and assessments to inform conservation planning and policy.

Systematic evidence map of research on life-history traits in native and introduced earthworms

This study maps research on life-history traits of native and introduced earthworms, finding that introduced species are more studied and research focuses on a few model species and specific environmental factors.

NANHA - Neurostimulation in Atypical Neurodevelopment: a Harmonized Atlas

NANHA is a curated database of non-invasive brain stimulation studies in neurodevelopmental disorders, providing standardized data for analysis and protocol development.

Benchmarking of bulk transcriptomic harmonization tools in a multi-platform B-cell lymphoma cohort identifies feature-specific quantile normalization and surrogate variable analysis as top-performing methods

Across 7,174 multi-platform lymphoma transcriptomes, only surrogate variable analysis and feature-specific quantile normalization corrected batch effects while keeping subtype and normal B-cell differences; top-scoring MNN erased biomarker correlations.

Evaluation of methods for AlphaFold-based integrative modeling

This study evaluates AlphaFold-based integrative modeling methods using chemical crosslinks, finding they often rely more on learned priors than input data.

The Unreasonable Effectiveness of Cell Types in Describing Neuronal Physiological Features

On 495 human Patch-seq neurons, cell type means predicted electrophysiology better than neural networks on scGPT embeddings or gene panels; fusing scGPT embeddings with cell type labels predicted best.

QuantiTrack: A unified software to study protein dynamics in living cells

QuantiTrack, a MATLAB GUI, unifies single-molecule detection, tracking, quality control and mobility analyses; for glucocorticoid receptor, hormone washout cut bound fraction, dwell times and low-mobility occupancy.

Accounting for social feasibility in landscape-scale scenarios for restoring nature beyond protected areas

The paper compares two restoration scenarios in the French Alps, showing that social feasibility and biophysical effectiveness often conflict, with spatial allocation of interventions being key to successful implementation.

BGC-QUAST: a quality assessment tool for genome mining software

BGC-QUAST evaluates and compares BGC predictions across samples, assemblies, and tools to facilitate biosynthetic potential analysis.

Benchmarking CUT&RUN analysis using motif enrichment

This study benchmarks CUT&RUN peak callers using motif enrichment, finding that MACS2 outperforms SEACR in most scenarios with optimal preprocessing.

Benchmarking Twist Genotyping-by-Sequencing Against Whole-Genome Sequencing in Nuclear Families

This study benchmarks Twist Genotyping-by-Sequencing against whole-genome sequencing in nuclear families, finding GxS slightly worse than GSA but still highly concordant with WGS.

Open-access template and database approaches for pseudo-CT generation in brain PET/MRI attenuation correction

The paper presents an open-access framework for PET attenuation correction using pseudo-CT methods with a database of MRI-CT scans, showing that UCL achieves the lowest quantification error.

A reproducible buccal transcriptomic response to music exposure in autism spectrum disorder: insights into the oral-brain axis

The paper identifies a reproducible transcriptional response to music exposure in ASD using buccal samples, highlighting neuroimmune pathways and a two-gene signature for monitoring interventions.

Estrogen interactions with breast cancer risk variants in regulatory DNA

The paper identifies estrogen-modulated SNVs associated with breast cancer risk and shows their interaction with reproductive factors using polygenic risk scores.

SenSASP: A Unified, Multi-Layer Database of Senescence and SASP Genes

Merging CellAge, GenAge, SenMayo and Reactome senescence genes onto Ensembl IDs yields 1,250 provenance-tracked genes, most also annotated with mouse/zebrafish orthologs, tissue expression and STRING interactions.

Revisiting Stress Granule Transcriptomes Suggests Mitochondrial RNA Enrichment Despite Methodological Bias

This study reanalyzes SG transcriptomes to show mitochondrial RNA enrichment despite methodological biases in isolation techniques.

Systematic benchmarking of small variant calling pipelines for long-read RNA sequencing data

The paper benchmarks small variant calling pipelines for long-read RNA sequencing data, finding that sequencing quality is the main factor affecting performance, followed by variant caller and alignment strategy.

A conserved transcriptional axis links cortical organization, developmental timing and autism-associated perturbations

The paper identifies a conserved transcriptional axis in human and mouse cortex linking developmental timing, cortical organization, and autism-associated mutations.