SmileyLlama: Modifying Large Language Models for Directed Chemical Space Exploration

📅 2024-09-03
🏛️ arXiv.org
📈 Citations: 7
Influential: 0
📄 PDF
🤖 AI Summary
This work addresses the challenge of transforming general-purpose large language models (LLMs) into attribute-controllable molecular generators. We propose a lightweight adaptation paradigm that converts open-source Llama models into chemical language models (CLMs) via supervised fine-tuning (SFT) and direct preference optimization (DPO), enabling direct SMILES string generation conditioned on multidimensional physicochemical properties (e.g., logP, aqueous solubility). To our knowledge, this is the first empirical demonstration that an adapted general LLM achieves performance on multi-objective molecular generation tasks comparable to or exceeding that of domain-specific chemically pretrained models. The approach enables a paradigm shift from “chemical knowledge question-answering” to “property-directed molecular design,” significantly enhancing controllability, interpretability, and interactive exploration of chemical space.

Technology Category

Application Category

📝 Abstract
Here we show that a Large Language Model (LLM) can serve as a foundation model for a Chemical Language Model (CLM) which performs at or above the level of CLMs trained solely on chemical SMILES string data. Using supervised fine-tuning (SFT) and direct preference optimization (DPO) on the open-source Llama LLM, we demonstrate that we can train an LLM to respond to prompts such as generating molecules with properties of interest to drug development. This overall framework allows an LLM to not just be a chatbot client for chemistry and materials tasks, but can be adapted to speak more directly as a CLM which can generate molecules with user-specified properties.
Problem

Research questions and friction points this paper is trying to address.

Transforming general LLM into chemical language model
Generating valid novel drug-like molecules efficiently
Optimizing molecules for 3D conformation binding affinity
Innovation

Methods, ideas, or system contributions that make the work stand out.

Supervised fine-tuning transforms LLM into CLM
Direct preference optimization enhances molecule generation
iMiner framework predicts drug molecules with 3D conformations
J
Joseph M. Cavanagh
Kenneth S. Pitzer Theory Center and Department of Chemistry, University of California, Berkeley
K
Kunyang Sun
Kenneth S. Pitzer Theory Center and Department of Chemistry, University of California, Berkeley
A
Andrew Gritsevskiy
Department of Computer Science, University of Wisconsin–Madison
D
Dorian Bagni
Kenneth S. Pitzer Theory Center and Department of Chemistry, University of California, Berkeley
T
Thomas D. Bannister
Department of Molecular Medicine, The Herbert Wertheim UF Scripps Institute for Biomedical Innovation and Technology
T
Teresa Head-Gordon
Departments of Bioengineering and Chemical and Biomolecular Engineering, University of California, Berkeley; Chemical Sciences Division, Lawrence Berkeley National Laboratory