MAHLER: New Machine Learning Method Predicts Antibody-Antigen Binding Kinetics at Scale
Researchers have developed MAHLER, an open-source machine learning and physics hybrid tool that predicts how long antibodies remain bound to their antigens, a property called residence time. Current computational antibody design tools focus mainly on binding affinity, but binding kinetics—how fast a drug binds and unbinds—are critical for real-world drug efficacy and pharmacokinetics. MAHLER achieves screening-grade accuracy in roughly 4 minutes per prediction on a single GPU, compared to days required by existing enhanced molecular dynamics methods.
MAHLER (Metadynamics-Anchored Hybrid Learning for Engineering off-Rates) combines metadynamics-based molecular dynamics simulations with inverse-folding machine learning models to predict relative antibody-antigen dissociation kinetics. The tool targets a gap in computational antibody engineering: while many existing methods optimize binding affinity, they largely ignore binding kinetics, which govern how long a drug remains active at its target and influence pharmacokinetic behavior in vivo. MAHLER demonstrates first-in-class screening-grade accuracy when predicting relative residence times across families of point mutants of an antibody-antigen complex. After an initial antigen-specific setup phase, each subsequent prediction requires only about 4 minutes on a single NVIDIA A100 GPU, a dramatic reduction from the days typically needed even with accelerated molecular dynamics approaches. The method is fully open-source, making it accessible to the broader antibody engineering and drug discovery community. The authors position MAHLER as a practical, kinetics-aware complement to affinity-focused design pipelines rather than a replacement for them.
What's missing
The study benchmarks MAHLER on point mutants within antibody families, leaving open how well it generalizes to structurally diverse antibody scaffolds or entirely novel antigen targets. The computational cost and complexity of the required initial antigen-specific setup phase are not fully detailed. It is also unclear how screening-grade accuracy was defined or what the false-positive and false-negative rates are in a realistic drug discovery campaign context.
What different sources said
- bioRxivCenter
MAHLER: Integrating Metadynamics and Inverse Folding to Predict Antibody-Antigen Kinetics
Related
Multiscale Brain Model Predicts Novel Propofol Anesthesia Biomarker Without Training on Clinical Data
Researchers developed a mechanistic computational model of thalamocortical brain circuits that successfully predicted a previously unnoticed dose-dependent biomarker of propofol anesthesia. The model, driven solely by GABA-A receptor modulation, reproduced empirical data from both macaques and humans without being fitted to any anesthesia-specific data. The findings suggest that simulation-first approaches could accelerate biomarker discovery in neuropharmacology without requiring large clinical datasets.
Green-Synthesized Zinc Oxide Nanoparticles from Mimosa pudica Show Biocompatibility with Bone Marrow Stem Cells in Lab Study
Researchers synthesized zinc oxide nanoparticles using Mimosa pudica leaf extract and tested their effects on human bone marrow mesenchymal stromal cells, finding the nanoparticles preserved cell viability, structure, and bone-forming capacity. The plant-derived nanoparticles outperformed both the raw plant extract and conventionally synthesized zinc oxide in maintaining cell metabolic activity over five days. The findings suggest these bioactive nanomaterials could be candidates for musculoskeletal tissue engineering, though the research remains at an early in vitro stage.
Study Compares Genetic Modeling Approaches for Dyadic Social Interactions in Animals
A new preprint study compared two statistical modeling approaches for analyzing the genetic basis of social interactions in animals, finding that dyadic models outperform marginal models that aggregate individual-level data. The research used pig aggression data from 797 finishing pigs across 59 social groups as a test case. The findings have implications for how animal geneticists model and interpret the heritable components of social behavior.