MAGNet: Motif-Agnostic Generation of Molecules from Scaffolds
BP2-STS
- Hetzel, Leon University of Munich
- Sommer, Johanna University of Munich
- Rieck, Bastian ORCID University of Fribourg
- Theis, Fabian Unviersity of Munich
- Günnemann, Stephan Unviersity of Munich
- 2025
Published in:
- The Thirteenth International Conference on Learning Representations (ICLR 2025 Spotlight). - 2025
English
Recent advances in machine learning for molecules exhibit great potential for facilitating drug discovery from in silico predictions.
Most models for molecule generation rely on the decomposition of molecules into frequently occurring substructures (motifs), from which they generate novel compounds.
While motif representations greatly aid in learning molecular distributions, such methods fail to represent substructures beyond their known motif set, posing a fundamental limitation for discovering novel compounds. To address this limitation and enhance structural expressivity, we propose to separate structure from features by abstracting motifs to scaffolds and, subsequently, allocating atom and bond types. To this end, we introduce a novel factorisation of the molecules' data distribution that considers the entire molecular context and facilitates learning adequate assignments of atoms and bonds to scaffolds. Complementary to this, we propose MAGNet, the first model to freely learn motifs. Importantly, we demonstrate that MAGNet's improved expressivity leads to molecules with more structural diversity and, at the same time, diverse atom and bond assignments.
Most models for molecule generation rely on the decomposition of molecules into frequently occurring substructures (motifs), from which they generate novel compounds.
While motif representations greatly aid in learning molecular distributions, such methods fail to represent substructures beyond their known motif set, posing a fundamental limitation for discovering novel compounds. To address this limitation and enhance structural expressivity, we propose to separate structure from features by abstracting motifs to scaffolds and, subsequently, allocating atom and bond types. To this end, we introduce a novel factorisation of the molecules' data distribution that considers the entire molecular context and facilitates learning adequate assignments of atoms and bonds to scaffolds. Complementary to this, we propose MAGNet, the first model to freely learn motifs. Importantly, we demonstrate that MAGNet's improved expressivity leads to molecules with more structural diversity and, at the same time, diverse atom and bond assignments.
- Faculty
- Faculté des sciences et de médecine
- Department
- Département d'Informatique
- Language
-
- English
- Classification
- Computer science and technology
- Other electronic version
- License
- CC BY
- Open access status
- diamond
- Persistent URL
- https://folia.unifr.ch/unifr/documents/331154
Statistics
Document views: 57
File downloads:
- 7094_magnet_motif_agnostic_gen.pdf: 51