benchmark study

Rodent chronic variable stress procedures: a disjunction between stress entity and impact on behaviour

This study examines how variations in chronic variable stress procedures affect behavioral outcomes in mice, finding that longer and more diverse stress regimens reduce effect sizes in certain tests.

Assessing coverage of the Monitoring Framework of the Kunming-Montreal Global Biodiversity Framework and opportunities to fill gaps.

This paper assesses the coverage of the GBF Monitoring Framework and identifies gaps and opportunities for improvement in biodiversity monitoring.

The FAIR Data Point Populator: collaborative FAIRification and population of FAIR Data Points

The paper introduces a tool for FAIR metadata creation that improves collaboration, schema adherence, and scalability compared to existing tools.

Benchmarking methods for mapping functional connectivity in the brain

This study benchmarks 239 pairwise interaction statistics for functional connectivity mapping, finding substantial variation in their topological, neurobiological, and behavioral properties.

Benchmarking Inverse Folding Models for Antibody CDR Sequence Design

This study benchmarks AI models for antibody CDR sequence design, finding that specialized models like AntiFold perform better in complex regions while general models struggle with antibody-specific nuances.

OriGene: A Self-Evolving Virtual Disease Biologist Automating Therapeutic Target Discovery

OriGene identifies therapeutic targets by integrating diverse data modalities and outperforms human experts and models in accuracy and robustness.

Polygenic modifiers of expressivity in telomere biology disorders

This study identifies polygenic modifiers of expressivity in telomere biology disorders by showing how common and rare genetic variants interact to influence clinical variability.

Structural and temporal dynamics analysis on PANoptosis in sepsis: a bibliometric analysis

This study performs a bibliometric analysis of PANoptosis in sepsis research from 2000 to 2024, identifying key themes, trends, and emerging subfields.

Validating data from Multiplex Assays of Variant Effect (MAVEs): A CanVIG-UK National Survey of NHS Clinical Scientists

The paper surveys NHS clinical scientists on their acceptance of MAVE data validation, finding that trusted central bodies are needed for clinical adoption.

Detection of astrocyte epigenetic memory in in vitro systems, experimental autoimmune encephalomyelitis and multiple sclerosis samples

The paper rebuts claims that astrocyte epigenetic memory is due to immune cell contamination, showing its existence in pure astrocyte cultures and MS samples.

Biomni: A General-Purpose Biomedical AI Agent

An LLM agent with a literature-mined environment of 150 tools, 105 packages and 59 databases plans in code, matching reported human accuracy on database and sequence question answering.

Finding an optimal sequencing strategy to detect short and long genetic variants in a human genome

This paper benchmarks sequencing technologies and variant callers for detecting SNPs, indels, and SVs, finding that SRS is optimal for small variants while LRS excels in SV detection and complex regions.

Comparison of sequence- and structure-based antibody clustering approaches on simulated repertoire sequencing data

This study compares sequence- and structure-based antibody clustering methods on simulated repertoire data, finding structure-based methods identify more multiple-occupancy clusters but have length constraints.

OpenASO: RNA Rescue--designing splice-modulating antisense oligonucleotides through community science

This study uses community science and RNA structure prediction to design splice-modulating ASOs that enhance exon 16 splicing in a hemophilia A model.

From Hype to Health Check: Critical Evaluation of Drug Response Prediction Models with DrEval

This paper evaluates drug response prediction models and finds that deep learning models perform similarly to simple baselines, highlighting issues in current model evaluation practices.

Benchmarking methods integrating GWAS and single-cell transcriptomic data for mapping trait-cell type associations

This paper benchmarks 20 methods for mapping trait-cell type associations by integrating GWAS and scRNA-seq data, finding CATCH to be the most robust approach.

FoldBench: An All-atom Benchmark for Biomolecular Structure Prediction

FoldBench evaluates all-atom biomolecular structure prediction models across nine tasks, revealing performance gaps and highlighting AlphaFold 3's superiority.

Fraudulent studies are undermining the reliability of systematic reviews - a study of the prevalence of problematic images in preclinical studies of depression

The paper examines the prevalence of fraudulent studies in a systematic review of depression research, finding that 19% of studies had problematic images, which biased review results and were not easily identifiable.

Assessing the Molecular Validity of Spontaneous Lupus Mouse Models and Its Implication for Human Studies

This study evaluates molecular pathways in four spontaneous lupus mouse models and compares them to human SLE to assess translational relevance and guide future research.

A curated collection of transcriptome datasets to study the transcriptional response in blood and nasal samples following viral respiratory inoculation and vaccination

The paper curates a collection of transcriptome datasets from viral respiratory inoculation and vaccination studies to enable reanalysis of transcriptional responses in blood and nasal samples.

PhylloTraits 1.0: Unveiling the diversity of functional traits

PhylloTraits 1.0 compiles functional trait data for 230 Phyllostomidae bat species to enhance understanding of their ecology, evolution, and conservation.

Switching ON Hydrogen Sulfide: A Chemogenetic Toolkit for Spatially Resolved H2S Manipulation

This paper benchmarks two enzymes for controlled H2S production, finding that stDCyD generates only H2S while mDAAO also produces hydrogen peroxide.

A computational strategy to uncover fusion genes in prostate cancer cell lines

This paper evaluates three fusion gene prediction tools for detecting TMPRSS2::ERG in prostate cancer cell lines using RNA sequencing data.

Artificial Intelligence's Contribution to Biomedical Literature Search: Revolutionizing or Complicating?

This study evaluates ChatGPT's utility for biomedical literature search compared to traditional methods, finding limitations in accuracy and consistency but potential with user-support tools.

Evaluating Foundation Models for In-Silico Perturbation

This paper evaluates foundation models for in-silico perturbation using a novel framework, revealing differences in their performance across biological contexts and gene categories.

Forskolin as a multi-target modulator of the oncogenic PI3K/Akt signaling pathway: An in silico integrative GO/KEGG enrichment analysis and network pharmacology

This study uses in silico methods to analyze Forskolin's anticancer mechanisms, revealing its targeting of the PI3K/Akt pathway and related oncogenic networks.

Linking international registries to FHIR and Phenopackets with RareLink: a scalable REDCap-based framework for rare disease data interoperability

RareLink enables interoperability between REDCap and global health standards by exporting data to Phenopackets and FHIR, improving rare disease research scalability and data reuse.

Defective But Promising: Evaluating Bioinformatic Pipelines for Utility of Defective Interfering RNA Discovery in Plant Viral Infections

The paper evaluates bioinformatic tools for identifying defective interfering RNA in plant viral infections and finds low agreement among tools but supports the prevalence of DI RNA under permissive conditions.

Introducing Large Language Models to Human-Based Etymological Classification in Zooplankton

The paper benchmarks large language models against expert classification of zooplankton species names, finding high agreement except for culturally nuanced terms.

Investigating the performance of Oxford Nanopore long-read sequencing with respect to Illumina microarrays and short-read sequencing

This study compares Oxford Nanopore long-read sequencing with Illumina short-read sequencing and microarrays across 14 human genomes, finding LRS has competitive accuracy for SNVs and superior performance for structural variants.