Related Experiment Videos
Toward Lab-Ready AI Synthesis Plans with Protection Strategies and Route Scoring
Annie M Westerlund1, Lukas M Sigmund1, Marco V Mijangos2
1Molecular AI, Discovery Sciences, R&D, AstraZeneca Gothenburg, Pepparedsleden 1, 43183 Mölndal, Sweden.
Journal of Chemical Information and Modeling
|May 23, 2026
Summary
AI tools for molecule synthesis planning now feature automated protection strategies. This innovation addresses selectivity issues in reaction trees, improving efficiency for chemists and accelerating drug discovery.
Area of Science:
- Chemical synthesis
- Computational chemistry
- Machine learning in chemistry
Background:
- AI-driven synthesis planning tools aid novel molecule creation.
- These tools often generate reaction trees with selectivity issues due to competing reactive sites.
- Manual postprocessing for protection strategies hinders full automation and efficiency.
Purpose of the Study:
- To develop automated routines for identifying competing sites and formulating context-aware protection strategies.
- To enhance AI-driven synthesis planning by addressing selectivity issues and improving route quality.
- To accelerate industrial drug discovery workflows through improved synthesis planning.
Main Methods:
- Leveraging machine learning and encoded human chemical knowledge (rules and data).
- Considering cross-functional-group competition, protecting group selection, and reaction tree structure.
- Implementing orthogonal and multistep protection strategies.
- Proposing a competing sites score for analyzing and reranking reaction trees.
Main Results:
- Substantial reduction in selectivity issues in AiZynthFinder-generated reaction trees.
- Improved quality of synthesis routes provided to users.
- Modest increase in computation time per target molecule.
- Demonstrated effectiveness of context-aware protecting group strategies.
Conclusions:
- Automated selectivity control via context-aware protecting group strategies enhances AI-driven synthesis planning.
- The framework significantly improves the efficiency and quality of AI-generated synthesis routes.
- This approach facilitates the acceleration of industrial drug discovery processes.
Related Concept Videos
01:16Non-equilibrium in the Cell
An important concept in studying metabolism and energy is that of chemical equilibrium. Most chemical reactions are reversible. They can proceed in both directions, releasing energy into their environment in one direction, and absorbing it from the environment in the other direction. The same is true for the chemical reactions involved in cell metabolism, such as the breaking down and building up of proteins into and from individual amino acids, respectively. Reactants within a closed system...
01:28ATP and Macromolecule Synthesis
Biological macromolecules are organic compounds, predominantly composed of carbon atoms. The carbon atoms are covalently bonded with hydrogen, oxygen, nitrogen, and other minor elements. There are four major biological macromolecule classes: carbohydrates, lipids, proteins, and nucleic acids.
Most macromolecules are composed of single subunits, or building blocks, called monomers. The monomers combine with each other using covalent bonds to form larger molecules known as polymers.
Conversion of...
Most macromolecules are composed of single subunits, or building blocks, called monomers. The monomers combine with each other using covalent bonds to form larger molecules known as polymers.
Conversion of...