Fraglingo: Molecular Design via Attachment-Aware Autoregressive Fragment Generation

• URL: http://arxiv.org/abs/2609.13519v1
• Date: Fri, 11 Sep 2026 20:42:04 GMT
• Title: Fraglingo: Molecular Design via Attachment-Aware Autoregressive Fragment Generation
• Abstract summary: We introduce Fraglingo, an autoregressive fragment-based molecular generator.<n>Fraglingo predicts an attachment-aware fragment embedding and retrieves the next fragment.<n>It generalizes to fragment libraries up to 4x larger than those used during training without retraining.
• Score: 58.88304971824104
• License: http://creativecommons.org/licenses/by/4.0/
• Abstract: Molecular design is most effective when generation mirrors the edits chemists actually make: extending a scaffold, replacing a substituent, or decorating a scaffold at a specified attachment site while optimizing molecular properties. Fragment-based molecular design naturally supports this workflow, yet existing approaches often separate fragment selection from attachment prediction, first choosing a fragment from a fixed vocabulary and then predicting how it should be connected. This decoupling restricts generation to a closed fragment vocabulary and treats attachment as a separate prediction problem. We introduce Fraglingo, an autoregressive fragment-based molecular generator that jointly models fragment identity and attachment in a continuous latent space. Fraglingo predicts an attachment-aware fragment embedding and retrieves the next fragment through latent-space nearest-neighbor search. To encode attachment context, we introduce a wildcard-anchored readout that represents the growing molecule from the perspective of its active attachment site, enabling the predicted embedding to capture both the molecular context and the required attachment. Because generation operates in a continuous embedding space rather than over fixed fragment identifiers, new fragments can be added to the inference-time vocabulary without retraining, provided their embeddings can be computed by the trained fragment encoder. This retrieval-based formulation provides a unified generation primitive for molecule generation, scaffold generation, scaffold decoration, and molecular optimization. On controlled property-conditional benchmarks, Fraglingo achieves stronger joint property control than comparably trained baselines while maintaining competitive validity, uniqueness, and novelty. Furthermore, Fraglingo generalizes to fragment libraries up to 4x larger than those used during training without retraining.

Related papers

• MolLingo: Molecule-Native Representations for LLM-Powered Scientific Agents [57.28652712579434]
  We present MolLingo, a multi-agent system that emulates the reasoning process of a chemist to automate molecular design.<n>MolLingo addresses this by coordinating a Literature Agent, a Chemist Agent, and an Orchestrator through a shared memory module, with each agent equipped with domain-specific tools.<n>As a case study in early-stage therapeutic design, MolLingo grounds the Chemist Agent's reasoning in binding site geometry and residue-level protein context.
  arXiv  Detail & Related papers  (2026-05-27T02:11:23Z)
• FORGE: Fragment-Oriented Ranking and Generation for Context-Aware Molecular Optimization [38.801595580724616]
  FORGE is a two-stage framework that reformulates molecular optimization as context-aware local editing.<n>Built on a compact 0.6B language model, FORGE consistently outperforms prior methods.<n>These results highlight the value of explicit fragment-level supervision as a more easily obtainable, scalable, and hallucination-less alternative to natural language training.
  arXiv  Detail & Related papers  (2026-05-11T09:06:42Z)
• Flow-Based Fragment Identification via Binding Site-Specific Latent Representations [7.729885628667701]
  We develop a protein-fragment encoder that relies on a contrastive learning approach to map both molecular fragments and protein surfaces in a shared latent space.<n>The encoder captures interaction-relevant features and allows to perform virtual screening as well as generative design with our new method LatentFrag.<n>Our generative method outperforms common methods such as virtual screening at a fraction of its computational cost providing a valuable starting point for fragment hit discovery.
  arXiv  Detail & Related papers  (2025-09-16T16:20:45Z)
• FragmentNet: Adaptive Graph Fragmentation for Graph-to-Sequence Molecular Representation Learning [0.09986418756990159]
  We introduce FragmentNet, a graph-to-sequence foundation model with an adaptive, learned tokenizer that decomposes molecular graphs into chemically valid fragments.<n>FragmentNet integrates VQVAE-GCN for hierarchical fragment embeddings, spatial positional encodings for graph serialization, global molecular descriptors, and a transformer.
  arXiv  Detail & Related papers  (2025-02-03T09:21:49Z)
• Molecule Generation with Fragment Retrieval Augmentation [41.95947899013865]
  Fragment Retrieval-Augmented Generation (f-RAG) is based on a pre-trained molecular generative model that proposes additional fragments to complete and generate a new molecule. To extrapolate beyond the existing fragments, f-RAG updates the fragment vocabulary with generated fragments via an iterative refinement process.
  arXiv  Detail & Related papers  (2024-11-18T21:43:52Z)
• FragRel: Exploiting Fragment-level Relations in the External Memory of Large Language Models [54.13671100638092]
  We propose a fragment-connected Hierarchical Memory based Large Language Models (LLMs) We formulate the fragment-level relations in external memory and present several instantiations for different text types. We validate the benefits of involving these relations on long story understanding, repository-level code generation, and long-term chatting.
  arXiv  Detail & Related papers  (2024-06-05T09:31:37Z)
• Drug Discovery with Dynamic Goal-aware Fragments [76.10700304803177]
  We propose a molecular generative framework for drug discovery, named Goal-aware fragment Extraction, Assembly, and Modification (GEAM) GEAM consists of three modules, each responsible for goal-aware fragment extraction, fragment assembly, and fragment modification. We experimentally demonstrate that GEAM effectively discovers drug candidates through the generative cycle of the three modules.
  arXiv  Detail & Related papers  (2023-10-02T01:30:42Z)
• De Novo Molecular Generation via Connection-aware Motif Mining [197.97528902698966]
  We propose a new method, MiCaM, to generate molecules based on mined connection-aware motifs. The obtained motif vocabulary consists of not only molecular motifs (i.e., the frequent fragments), but also their connection information. Based on the mined connection-aware motifs, MiCaM builds a connection-aware generator, which simultaneously picks up motifs and determines how they are connected.
  arXiv  Detail & Related papers  (2023-02-02T14:40:47Z)
• Equivariant 3D-Conditional Diffusion Models for Molecular Linker Design [82.23006955069229]
  We propose DiffLinker, an E(3)-equivariant 3D-conditional diffusion model for molecular linker design. Our model places missing atoms in between and designs a molecule incorporating all the initial fragments. We demonstrate that DiffLinker outperforms other methods on the standard datasets generating more diverse and synthetically-accessible molecules.
  arXiv  Detail & Related papers  (2022-10-11T09:13:37Z)

This list is automatically generated from the titles and abstracts of the papers in this site.

This site does not guarantee the quality of this site (including all information) and is not responsible for any consequences.