The Genetic Engineering Gazette

Nov 2 – Nov 8, 2026

A unified coarse-grained neural network potential for the dynamics of biomolecular complexes

AMARO v2.0 is a simulation method that models proteins and drugs using only their heavier atoms, so their movements can be studied faster. Faster simulations could help researchers understand biological processes and speed up drug discovery.

A unified coarse-grained neural network potential for the dynamics of biomolecular complexes · Antonio Mirarchi, Stephen E. Farr, Adrià Peréz et al. · arXiv:2610.37606

Accurate structural modelling of protein-glycan recognition with GlycoWeave

GlycoWeave is an AI method that predicts how proteins bind to complex sugar molecules called glycans, outperforming general models like AlphaFold 3. It could help researchers understand diseases and develop drugs that target protein-sugar interactions.

Accurate structural modelling of protein-glycan recognition with GlycoWeave · Huifeng Zhao, Jintu Zhang, Yifei Xiao et al. · arXiv:2610.37410

Evolutionary characterization of whole-genome doubling across human cancers

Cancer cells that duplicate their whole chromosome set later lose chromosomes less specifically and show reduced immune responses through changes in gene regulation. It offers insights into how cancers progress and evade the immune system, which could inform future treatments.

Evolutionary characterization of whole-genome doubling across human cancers · Yangyang Zhang, Haiming Li, Xiaoyu Zhao et al. · arXiv:2611.01897

Generative discovery of chemical reaction networks

ReactNexus is an AI system that discovers chemical reaction pathways and networks starting only from initial molecules and no known products. It could help scientists understand reactions, design more efficient processes and discover new compounds faster.

Generative discovery of chemical reaction networks · Yunhong Han, Gaopeng Ren, Yan Liu et al. · arXiv:2610.38236

RangeFinder: Predicting Ligand Property Ranges from Protein Pockets

RangeFinder is an AI model that reads a protein's binding site and predicts the range of properties suitable drug-like molecules should have. It offers a fast way to narrow many candidate molecules, helping researchers focus efforts in early drug discovery.

RangeFinder: Predicting Ligand Property Ranges from Protein Pockets · Alan Nadelsticher Ruvalcaba, Chris Alvin, Supratik Mukhopadhyay et al. · arXiv:2611.03379

A Scientific Claim Stability Framework for Evaluating MRI Morphometry Pipelines

The framework checks whether conclusions about aging or disease from MRI brain measurements stay consistent when methods or datasets change. It helps researchers identify robust findings, supporting solid, reproducible evidence about the brain.

A Scientific Claim Stability Framework for Evaluating MRI Morphometry Pipelines · Duy-Cat Can, Dang-Duat Tran, Anh-Khang Nguyen et al. · arXiv:2611.02413

An auditable agentic pipeline for variant-to-gene prioritisation, validated against orthogonal functional evidence

V2G, an automated system that helps pinpoint which specific gene is affected by a genetic variant found in studies like GWAS. Crucially, it uses a large language model to re-rank candidate genes, improving the accuracy of identifying the correct one.

An auditable agentic pipeline for variant-to-gene prioritisation, validated against orthogonal functional evidence · Shakuntala Baichoo, Sohail, Sarkar et al. · arXiv:2611.01461

OpenULF: A Modular Toolkit Enables Data-Driven Ultra-Low-Field MRI with both Classical and Quantum Sensors

OpenULF is an open-source simulation toolkit that generates realistic data to train AI models for ultra-low-field MRI. It supports work on portable, low-cost MRI alternatives and shows sensor choice depends on the medical application and field strength.

OpenULF: A Modular Toolkit Enables Data-Driven Ultra-Low-Field MRI with both Classical and Quantum Sensors · Hanhan Wu, Yuchen Yuan, Weiwen Jiang · arXiv:2611.01511

Detection Instead of Inversion: T1-T2 Microstructure Quantification With Detection Transformers

The method analyzes specialized MRI data by treating the task like object detection, directly pinpointing and quantifying tiny tissue compartments. It offers a faster, more precise tool to study biological tissues without assuming their underlying structure.

Detection Instead of Inversion: T1-T2 Microstructure Quantification With Detection Transformers · Sebastian Endt, Fatih Özkan, Marion I. Menzel · arXiv:2611.01790

All Gazettes · Genetic Engineering & Bio-medical Science