benchmark study

Shared brain basis for altered self-referential processing across psychiatric disorders? A systematic review and meta-analysis of neuroimaging studies

This study reviews and meta-analyzes neuroimaging studies on self-referential processing in psychiatric disorders, finding shared neural aberrancies but limited evidence for a transdiagnostic feature.

An explanatory benchmark of spatial domain detection reveals key drivers of method performance

This paper benchmarks 26 spatial domain detection methods across real and semi-synthetic spatial transcriptomics data to identify performance drivers and introduce a modular benchmarking framework.

Comparison of state-of-the-art error-correction coding for sequence-based DNA data storage

This study benchmarks six DNA data storage codecs using standardized in silico and in vitro experiments, finding common benchmarks inadequate and demonstrating data storage capacities of 43 EB g-1 and 13 EB g-1.

Multi-modal benchmarking of the Ultima UG100 and Illumina NovaSeq sequencing platforms using clinically relevant FFPE tissues

This paper benchmarks the Ultima UG100 and Illumina NovaSeq platforms on FFPE tissues for various sequencing modalities, finding they produce comparable results with subtle discrepancies in non-coding transcripts.

Benchmarking zero-shot single-cell foundation model embeddings for cellular dynamics reconstruction

This study benchmarks zero-shot single-cell foundation model embeddings for cellular dynamics reconstruction and finds they underperform a traditional HVG baseline in capturing complex temporal patterns.

Benchmarking BEAGLE to find optimal parameters for BEAST X

The paper benchmarks BEAGLE integration in BEAST X to determine optimal parameters for different hardware configurations using Dengue Virus data and simulations.

Benchmarking DNA Foundation Models: Biological Blind Spots in Evo2 Variant-Effect Prediction

This study benchmarks DNA foundation models like Evo2, revealing blind spots in variant-effect prediction and highlighting reliance on sequence regularities over biological constraints.

MESSI: Multimodal Experiments with SyStematic Interrogation using nextflow

MESSI benchmarks multimodal integration methods using standardized workflows and cross-validation, showing method performance varies by dataset and task.

invertmeeg: A Benchmark and Unified Python Library for EEGInverse Solvers

The paper introduces a benchmark and Python library for EEG inverse solvers, showing no single method dominates all scenarios and highlighting the performance of flexible subspace and hybrid methods.

EEG correlates of auditory rise time processing: A systematic review

This review systematically examines EEG markers of auditory rise time processing, finding that prolonged rise time decreases ERP component amplitudes and increases latencies, with varying sensitivity across components and experimental conditions.

Benchmarking tissue- and cell type-of-origin deconvolution in cell-free transcriptomics

This study benchmarks deconvolution methods for inferring tissue and cell type-of-origin from plasma cfRNA, showing method and reference variability impact performance.

CactEcoDB: Trait, spatial, environmental, phylogenetic and diversification data for the cactus family

CactEcoDB provides integrated ecological, trait, and phylogenetic data for over 1,000 cactus species to support comparative research on trait evolution and macroevolutionary dynamics.

TFBSpedia: a comprehensive human and mouse transcription factor binding sites database

The paper constructs a comprehensive TFBS database by integrating multiple resources and evaluates their accuracy using independent datasets.

Grounding olfactory perception in language: Benchmarks and models for generating natural language odor descriptions

ODIEU pairs human odor texts for 10,124 molecules with olfaction-tuned Sentence-BERT evaluators; CIRANO, a frozen GPT-2 with a molecular prefix, generates descriptions approaching human agreement.

Confidence Without Verification: Screening pLDDT Unreliability in AlphaFold2 Fold-Switching Predictions

The paper identifies unreliable confidence metrics in AlphaFold2 predictions for fold-switching proteins and introduces a screening method to detect false positives.

Automated high-throughput fabrication of patient-specific vessel-on-chips enables a generative AI digital twin--Cascade Learner of Thrombosis (CLoT) for personalized thrombosis prediction

The paper develops a platform for automated vessel-on-chip fabrication and a generative AI model, CLoT, to predict thrombosis in patients with high accuracy.

Why Boolean network control tools disagree: a taxonomy of control problems

The paper surveys Boolean network control tools, identifies inconsistencies in their results, and introduces a metric to prioritize mutations based on predicted phenotypic impact.

Structural Plausibility Without Binding Specificity: Limits of AI-Based Antibody-Antigen Structure Prediction Confidence Scores

The paper evaluates AI-based antibody-antigen structure prediction methods, finding that confidence scores often fail to distinguish correct from incorrect pairings, even with increased sampling.

Cross-ancestry performance of Parkinson's disease polygenic risk scores in admixed Latin American populations

This study evaluates polygenic risk scores for Parkinson's disease in admixed Latin American populations, finding that European-derived PRS outperform ancestry-matched but underpowered datasets.

Challenges and Opportunities in Single-Sample Network Modeling

This paper analyzes single-sample network inference methods, identifying trade-offs between accuracy and specificity and suggesting opportunities for methodological synergy.

The Effect of Neurodegeneration on Ultrasonic Vocalisations (USV) and Their Neuronal Substrates in Mice and Rats: A Systematic Review

This review examines how neurodegeneration in mice and rats affects ultrasonic vocalisations and their neuronal substrates, aiming to identify biomarkers for human speech deficits.

Whole-genome benchmarking reveals context-specific error rates in the Ultima UG100 and Illumina NovaSeqX Platforms.

This study benchmarks two whole-genome sequencing platforms, showing context-specific error rates and highlighting the importance of whole-genome benchmarking for accuracy assessment.

A Query-to-Dashboard Framework for Reproducible PubMed-Scale Bibliometrics and Trend Intelligence

PubMed Atlas provides a reproducible framework for analyzing PubMed-scale bibliometric data to identify scientific trends and research gaps.

Beyond alignment: synergistic integration is required for multimodal cell foundation models

This paper introduces a metric to evaluate synergistic integration in multimodal cell models and shows when it benefits tasks over unimodal approaches.

A Systematic Evaluation of Protein Phase Separation Predictors Across Diverse Protein Landscapes

The paper evaluates nine LLPS predictors across diverse protein landscapes, finding significant variability in performance and highlighting the need for diverse benchmarking data.

Adversarial Sequence Mutations in AlphaFold andESMFold Reveal Nonphysical StructuralInvariance, Confidence Failures, and Concerns forProtein Design

The paper evaluates AlphaFold and ESMFold by introducing adversarial mutations and finds structural invariance and unreliable confidence metrics, suggesting reliance on templates rather than biophysical principles.

Comparing Neanderthal introgression maps reveals core agreement but substantial heterogeneity

The paper compares Neanderthal introgression maps from 12 algorithms, finding core agreement but substantial heterogeneity, and recommends using multiple maps for robust conclusions.

MetaOmixTools: an interactive web suite for meta-analysis of ranked features and functional enrichment

MetaOmixTools provides a web-based platform for meta-analysis of ranked features and functional enrichment, enabling integration of multi-study omics data to identify conserved biological insights.

Critical Assessment of ML models for ADMET Prediction in TDC leaderboards

The paper evaluates ADMET prediction models on TDC leaderboards, finding most top models lack reproducibility and exhibit data leakage, with only three methods passing rigorous checks.

Benchmarking computational tools for locus-specific analysis of transposable elements in single-cell RNA-seq datasets

The paper benchmarks tools for locus-specific TE analysis in scRNA-seq, finding older TEs are more accurately quantified, while young TEs face challenges due to multi-mapping reads.