benchmarking

T2G: fast and accurate transcriptome to genome BAM liftover for short and long read RNA sequencing

T2G converts transcriptome-aligned BAM files to genome coordinates, improving specificity and splice junction accuracy across RNA sequencing technologies.

The Muskox Way: Humans, plants and animals in northeast Greenland through the Holocene

The paper uses sedimentary ancient DNA to analyze the timing of plant and animal arrivals in northeast Greenland, showing muskox arrived before humans and environmental changes coincided with cultural periods.

Genome-scale base editing reveals the functional effects of variants in yeast

This study uses genome-scale CRISPR base editing in yeast to assess the functional effects of 308,461 variants, identifying 18,009 with significant growth impacts.

afMLevel: Deep learning U-Net Models for Levelling Atomic Force Microscopy Images and Movies

This paper introduces two U-Net models for automatically levelling AFM images and movies, outperforming classical methods and reducing manual intervention.

A statistical framework for system-level synthesis

This paper introduces a statistical framework for system-level synthesis that learns and predicts biological processes from high-dimensional data while identifying participating features.

Differential sensitivity of k-mer based identification tools using closely related database sequences affects relative abundance estimation

This study evaluates how k-mer based tools affect abundance estimation in metagenomic data with closely related references, showing differential performance based on sample-reference distance and reference similarity.

Biome-specific genome catalogues reveal functional potential of shallow sequencing

The paper introduces BioSIFTR, a tool that uses shallow shotgun sequencing and biome-specific databases to generate accurate taxonomic and functional profiles, validated against deep sequencing data.

Sequencing-Free Protospacer-Adjacent Motif Discovery for Uncharacterized CRISPR-Cas9 Orthologs via Active Learning

This paper introduces an active learning method to discover PAMs for uncharacterized CRISPR-Cas9 orthologs without sequencing, reducing data requirements and validating it on known and novel Cas9 variants.

Water under the bridge: Aquatic environments and crepuscular-cathemeral intermediates facilitate temporal niche transitions in artiodactyls

The paper tests the crepuscular-cathemeral bridge hypothesis by comparing activity patterns in aquatic and terrestrial artiodactyls, finding support for the bridge and distinct roles of crepuscularity and cathemerality.

A genomic foundation model for interpreting human genetic diversity

The paper introduces hEVO, a genomic foundation model trained on human haplotype sequences, achieving top performance in variant interpretation and functional annotation across multiple benchmarks.

LLM-assisted analytical review catches statistical errors that human researchers miss: documented evidence from two computational biology case studies

Livestock depredation frequency is negatively associated with wolf body mass

The paper examines the relationship between wolf body mass and livestock depredation frequency, finding a negative association where higher body mass correlates with lower depredation rates.

NIH's dedicated K23 trial pathway was associated with more K-trial leadership but no detectable gain in new trials

This paper evaluates the impact of NIH's K23 trial pathway on trial leadership, finding increased K-trial leadership but no significant increase in new trials.

Functional architecture of rare protein coding variation across 786,871 exomes

The paper introduces KASM to integrate functional annotations and estimate heritability of rare protein coding variants across 786,871 exomes, finding higher heritability than previous estimates.

NED-Net: an annotation-training loop for rodent EEG event detection

NED-Net is an open-source tool for training and adapting seizure detectors in rodent EEG data, improving detection accuracy through iterative annotation and retraining.

Population genomics of Actinobacillus pleuropneumoniae reveals lineage-associated K:O antigen and Apx toxin diversity

The paper analyzes 328 Actinobacillus pleuropneumoniae genomes to explore relationships between genomic lineages, K:O antigen genotypes, and Apx toxin repertoires, finding strong associations between lineage and these traits.

Making the Impossible Possible: Uniting TMT and DIA for Scalable Low-Input-Proteomics

This paper combines TMT and DIA proteomics techniques to improve low-input proteomics, showing how to overcome co-isolation interference and enabling TMT quantification with DIA-focused software.

Rare variant burden complements polygenic risk in breast cancer and is associated with earlier diagnosis

This study shows that rare variant burden adds information beyond polygenic risk scores for breast cancer and is more strongly associated with earlier diagnosis.

Diversifications of both the three domains of life and SARS-CoV-2 possibly driven by biases between amino acid biosynthetic families

The paper proposes a mechanism for the diversification of life and SARS-CoV-2 based on biases between amino acid biosynthetic families, using proteomic features to support the three-domain classification.

S4-Oligo: optimizing polygenic score methods for clinical deployment

This paper introduces S4-Oligo, an optimized polygenic score method for clinical use, which matches or exceeds existing models' accuracy while using fewer SNPs.

Locally deployed AI agents for CRISPR guide discovery from outbreak reports

The paper develops a CRISPR guide-discovery pipeline using locally deployed AI agents and evaluates it with large language models.

Carbon-A: Unified Genome Annotation across Eukaryotic Diversity

Carbon-A annotates protein-coding genes across eukaryotic genomes using a large encoder model, achieving high accuracy and enabling the creation of a comprehensive annotation database.

Environmental drivers of soil virus diversity across UK habitats

This study identifies soil pH and habitat type as key drivers of viral diversity and community structure in UK soils, showing strong coupling with bacterial communities.

Hybrid Semantic-Lexical Retrieval for Source-to-CDE Mapping in the NCI caDSR

The paper develops a hybrid retrieval system for mapping source data elements to CDEs, achieving high recall across multiple datasets.

geoGraph: Connectivity modelling across geographic space with hexagonal discrete global grids

geoGraph models organism movement and connectivity across geographic space using hexagonal grids, improving accuracy over rectangular grids by avoiding distortion.

Integrating Multisource Target Evidence with Machine-Learning Potency Prediction for Alzheimer's Disease Drug Discovery

The paper develops a platform integrating multisource target evidence with machine learning for Alzheimer's drug discovery, improving potency prediction and model performance.

REM Sleep Iota Oscillations Across the Lifespan

This study examines iota oscillations in REM sleep across age groups, finding age-related changes in frequency, amplitude, and bandwidth, with high heritability for most properties.

The impact of strain heterogeneity on genomic profiling accuracy in mixed-strain Escherichia coli metagenomes

The paper evaluates how strain heterogeneity affects the accuracy of genomic profiling in mixed-strain Escherichia coli metagenomes using simulated data and various bioinformatic tools.

Validating OpenCap Kinematics for ACL Injury-Related Movements In and Out of the Lab

This study evaluates OpenCap's accuracy in measuring ACL injury risk factors both in and out of the lab compared to optical motion capture.

Simulation-free approximate Bayesian computation for stochastic reaction networks

This paper introduces SFABC, a simulation-free method for Bayesian inference of stochastic reaction networks, showing it matches simulation-based accuracy at lower computational cost.