Tag: computational biology

A multiscale simulation found that endothelial repair depended on early, moderate AKT activation and on balancing local nitric oxide production ...

HGACL-DRP achieved reported mean AUC values of 95.48% and 98.99% on two public cancer cell-line datasets in a computational drug-response ...

A 1,000-protein, five-organism benchmark found that leading open-source language models showed biological semantic alignment but achieved a best macro-F1 of ...

An interpretable deep learning model called SPACER reads spatial transcriptomics data to reveal the molecular and spatial rules that govern ...

A systematic reusability study of the CORNETO framework reveals that multi-sample biological network inference operates in distinct robustness regimes governed ...

A new open-source tool called ampliconaudit detects silent failures in amplicon taxonomic pipelines by auditing post-hoc artifacts, ensuring that computational ...

A new multi-evidence framework resolves coding-noncoding ambiguity in the Arrhenatherum elatius transcriptome, identifying 31.71% of transcripts as part of a ...

Researchers used RFdiffusion, ProteinMPNN, and molecular dynamics to design novel protein binders against the human sweet taste receptor TAS1R2, creating ...

A new two-step machine-learning framework called CLIMB and CLIFF computationally deconvolves bulk RNA-seq and drug sensitivity data into cell-type and ...

A new framework proposes translating natural-language prompts into machine-checkable constraints and a six-layer validator stack to make AI-driven genome design ...

Researchers propose a closed-loop framework that unites AI-driven sequence design, molecular dynamics simulation, and experimental validation to rationally engineer high-affinity, ...

A new study of immunized mice shows that compartment size and clonal proliferation, rather than T cell receptor sequence features, ...

A new multimodal AI framework called MEGH-DDI combines chemical language models with graph neural networks to predict drug–drug interactions with ...

Researchers have unveiled LoL-align, a probabilistic protein structure alignment algorithm that outperforms TM-align and Dali in detecting remote homologs and ...

Researchers have developed SpatialZoomer, a spectral graph-based tool that automatically identifies the biologically meaningful scales in spatial transcriptomics data, revealing ...

A new hybrid framework called HDOCK-Multimer combines protein-protein docking with combinatorial assembly to predict the structures of large protein complexes ...

A researcher at IIIT Allahabad has developed IDP-T5CKNN, a lightweight protein language model method that outperforms more than 20 existing ...

Researchers have developed extended phase mapping, a corrected cardiac mapping technique that explicitly accounts for phase defects and eliminates both ...

A Transformer deep learning model predicts cancer cell fates directly from live-cell videos with high accuracy up to eight hours ...

A systematic review of 38 studies finds that protein language models excel on benchmarks but largely lack experimental validation, prompting ...