BAGEL Enables Energy-Based Programmable Protein Engineering
Researchers publish BAGEL on December 3, 2025, a modular open-source Python framework that frames protein design as energy-function sampling and enables model-agnostic, gradient-free sequence exploration. BAGEL supports multi-state optimization, advanced Monte Carlo methods, and integration with public deep-learning folding and design models, and demonstrates four use cases including de novo peptide binders and enzyme variants. The release (v0.1) is available on GitHub and Zenodo.
Key Points
- 1Introduces BAGEL, a modular open-source framework for energy-function driven protein design.
- 2Enables model-agnostic, gradient-free exploration supporting multi-state optimization and advanced Monte Carlo.
- 3Allows practitioners to combine custom energy terms and plug deep-learning models for flexible design.
Scoring Rationale
High novelty and practical utility across protein design, balanced by initial v0.1 release and need for community validation.
Sources
Public references used for this report.
Practice interview problems based on real data
1,625 SQL & Python problems across 15 industry datasets — the exact type of data you work with.
Try 250 free problems