SynLlama: Generating Synthesizable Molecules and Their Analogs with Large Language Models

• URL: http://arxiv.org/abs/2503.12602v3
• Date: Fri, 18 Apr 2025 22:32:14 GMT
• Title: SynLlama: Generating Synthesizable Molecules and Their Analogs with Large Language Models
• Authors: Kunyang Sun, Dorian Bagni, Joseph M. Cavanagh, Yingze Wang, Jacob M. Sawyer, Andrew Gritsevskiy, Oufan Zhang, Teresa Head-Gordon,
• Abstract summary: We present a novel approach by fine-tuning Meta's Llama3 Large Language Models to create SynLlama.<n> SynLlama generates full synthetic pathways made of commonly accessible building blocks and robust organic reaction templates.<n>We find that SynLlama, even without training on external building blocks, can effectively generalize to unseen yet purchasable building blocks.
• Score: 2.4120602995529317
• License: http://creativecommons.org/licenses/by-sa/4.0/
• Abstract: Generative machine learning models for small molecule drug discovery have shown immense promise, but many molecules they generate are too difficult to synthesize, making them impractical for further investigation or development. In this work, we present a novel approach by fine-tuning Meta's Llama3 Large Language Models (LLMs) to create SynLlama, which generates full synthetic pathways made of commonly accessible building blocks and robust organic reaction templates. SynLlama explores a large synthesizable space using significantly less data compared to other state-of-the-art methods, and offers strong performance in bottom-up synthesis, synthesizable analog generation, and hit expansion, offering medicinal chemists a valuable tool for drug discovery developments. We find that SynLlama, even without training on external building blocks, can effectively generalize to unseen yet purchasable building blocks, meaning that its reconstruction capabilities extend to a broader synthesizable chemical space than the training data. We also demonstrate the use of SynLlama in a pharmaceutical context for synthesis planning of analog molecules and hit expansion leads for proposed inhibitors of target proteins.

Related papers

• SDDBench: A Benchmark for Synthesizable Drug Design [31.739548311094843]
  We propose a new, data-driven metric to evaluate molecule synthesizability. Our approach directly assesses the feasibility of synthetic routes for a given molecule through our proposed round-trip score. To demonstrate the efficacy of our method, we conduct a comprehensive evaluation of round-trip scores alongside search success rate across a range of representative molecule generative models.
  arXiv  Detail & Related papers  (2024-11-13T03:08:33Z)
• Generative Artificial Intelligence for Navigating Synthesizable Chemical Space [25.65907958071386]
  We introduce SynFormer, a generative modeling framework designed to efficiently explore and navigate synthesizable chemical space. By incorporating a scalable transformer architecture and a diffusion module for building block selection, SynFormer surpasses existing models in synthesizable molecular design.
  arXiv  Detail & Related papers  (2024-10-04T15:09:05Z)
• SynthFormer: Equivariant Pharmacophore-based Generation of Synthesizable Molecules for Ligand-Based Drug Design [19.578382119811238]
  We introduce SynthFormer, a novel machine learning model that generates fully synthesizable molecules, structured as synthetic trees, by introducing both 3D information and pharmacophores as input.<n>It is a first-of-its-kind approach that could provide capabilities for designing active molecules based on pharmacophores.
  arXiv  Detail & Related papers  (2024-10-03T17:38:46Z)
• Procedural Synthesis of Synthesizable Molecules [22.905205379063148]
  Design of synthetically accessible molecules and recommending analogs to unsynthesizable molecules are important problems for accelerating molecular discovery. We reconceptualize both problems using ideas from program synthesis. We create a bilevel framework for reasoning about the space of synthesis pathways.
  arXiv  Detail & Related papers  (2024-08-24T04:32:36Z)
• BatGPT-Chem: A Foundation Large Model For Retrosynthesis Prediction [65.93303145891628]
  BatGPT-Chem is a large language model with 15 billion parameters, tailored for enhanced retrosynthesis prediction. Our model captures a broad spectrum of chemical knowledge, enabling precise prediction of reaction conditions. This development empowers chemists to adeptly address novel compounds, potentially expediting the innovation cycle in drug manufacturing and materials science.
  arXiv  Detail & Related papers  (2024-08-19T05:17:40Z)
• RGFN: Synthesizable Molecular Generation Using GFlowNets [51.33672611338754]
  We propose Reaction-GFlowNet, an extension of the GFlowNet framework that operates directly in the space of chemical reactions. RGFN allows out-of-the-box synthesizability while maintaining comparable quality of generated candidates. We demonstrate the effectiveness of the proposed approach across a range of oracle models, including pretrained proxy models and GPU-accelerated docking.
  arXiv  Detail & Related papers  (2024-06-01T13:11:11Z)
• SynFlowNet: Design of Diverse and Novel Molecules with Synthesis Constraints [16.21161274235011]
  We introduce SynFlowNet, a GFlowNet model whose action space uses chemical reactions and buyable reactants to sequentially build new molecules. By incorporating forward synthesis as an explicit constraint of the generative mechanism, we aim at bridging the gap between in silico molecular generation and real world synthesis capabilities.
  arXiv  Detail & Related papers  (2024-05-02T10:15:59Z)
• Retrieval-based Controllable Molecule Generation [63.44583084888342]
  We propose a new retrieval-based framework for controllable molecule generation. We use a small set of molecules to steer the pre-trained generative model towards synthesizing molecules that satisfy the given design criteria. Our approach is agnostic to the choice of generative models and requires no task-specific fine-tuning.
  arXiv  Detail & Related papers  (2022-08-23T17:01:16Z)
• RetroGNN: Approximating Retrosynthesis by Graph Neural Networks for De Novo Drug Design [75.14290780116002]
  We train deep graph neural networks to approximate the outputs of a retrosynthesis planning software. Our approach finds molecules predicted to be more likely to be antibiotics while maintaining good drug-like properties and being easily synthesizable.
  arXiv  Detail & Related papers  (2020-11-25T22:04:16Z)
• RetroXpert: Decompose Retrosynthesis Prediction like a Chemist [60.463900712314754]
  We devise a novel template-free algorithm for automatic retrosynthetic expansion. Our method disassembles retrosynthesis into two steps. While outperforming the state-of-the-art baselines, our model also provides chemically reasonable interpretation.
  arXiv  Detail & Related papers  (2020-11-04T04:35:34Z)
• Learning To Navigate The Synthetically Accessible Chemical Space Using Reinforcement Learning [75.95376096628135]
  We propose a novel forward synthesis framework powered by reinforcement learning (RL) for de novo drug design. In this setup, the agent learns to navigate through the immense synthetically accessible chemical space. We describe how the end-to-end training in this study represents an important paradigm in radically expanding the synthesizable chemical space.
  arXiv  Detail & Related papers  (2020-04-26T21:40:03Z)
• The Synthesizability of Molecules Proposed by Generative Models [3.032184156362992]
  Discovery of functional molecules is an expensive and time-consuming process. One class of techniques of growing interest for early-stage drug discovery is de novo molecular generation and optimization. These techniques can suggest novel molecular structures intended to maximize a multi-objective function. However, the utility of these approaches is stymied by ignorance of synthesizability.
  arXiv  Detail & Related papers  (2020-02-17T15:41:28Z)

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.