bio.rodeo
ModelsOrganizationsProvidersLeaderboardAboutSign in
bio.rodeo

The authoritative source for evaluating biological foundation models. No hype, just honest analysis.

Categories
  • DNA & Gene
  • RNA
  • Protein
  • Small molecule
  • Single-cell
  • Spatial omics
  • Pathology
  • Imaging
  • Metabolomics
  • Biosignals
  • Language model
bio.rodeoModelsOrganizationsProvidersLeaderboardAboutFAQSubmit a modelTermsPrivacyContact
© 2026 Pulsatance. All rights reserved. ~
Built by Pulsatance
models / small-molecule / chatmol-numerical
Small moleculeLanguage model
Soochow UniversityReleased February 2025

ChatMol

Goal-oriented de novo molecule design with an instruction-tuned LLM that honors property targets, substructure constraints, and numeric values.

The short version

  • —Designs molecules to order from a written prompt naming the properties you want
  • —Numeric position encoding lets the model read a requested property value as a quantity
  • —Names a scaffold to keep in the same prompt that sets the property targets
  • —Property-prediction feedback pulls generated distributions toward the requested values
  • —One model covers single-property, substructure, and multi-property design tasks
10Openness

Where to run it

No providers recorded yet. Browse all providers

Goal-oriented molecule design asks for compounds that satisfy explicit constraints: a target binding affinity, a drug-likeness score above some threshold, a scaffold that must be retained. The dominant approaches — Bayesian optimization over a learned latent space, or reinforcement learning against a reward model — each require training property predictors, and both handle substructure requirements awkwardly, since a latent vector has no natural slot for "keep this ring system."

ChatMol, from researchers at Soochow University, reframes the problem as instruction following. A pretrained large language model is fine-tuned to read a natural-language prompt describing the desired properties and substructures and to emit a molecule that satisfies them. Two additions make this work in practice. First, the authors design a molecule string format derived from SELFIES, stripping the square brackets and separating elements with spaces, which keeps the validity guarantees of SELFIES while producing shorter, more language-like sequences that fit an existing LLM vocabulary without extending it. Second, they address a well-known weakness of language models — treating numbers as arbitrary tokens — by adding a dedicated positional encoding for numeric values in the prompt, so that a requested property value is represented as a quantity rather than as digits.

Note that an unrelated project also called ChatMol exists: a PyMOL plugin that acts as a conversational assistant for molecular modeling. It shares only the name.

#Key Features

  • Constraint-conditioned generation: A single fine-tuned model covers single-property, substructure-plus-property, and multi-property constrained design, where prior methods typically train a separate optimizer per objective.
  • Language-friendly molecule encoding: The bracket-free, space-separated SELFIES variant preserves syntactic validity while shortening sequences and reusing the base model's existing token vocabulary.
  • Numerical position encoding: Additional encoding applied to numeric values in the prompt gives the model a representation of magnitude rather than digit strings, raising the correlation between instructed and achieved property values.
  • Property-prediction feedback learning: Generated molecules are scored and the signal is fed back during fine-tuning, aligning the output distribution with the requested constraints without a full reinforcement learning loop.
  • Parameter-efficient adaptation: LoRA adapters keep the trainable footprint at roughly 21 million parameters, about 0.26% of the 8B backbone.

#Technical Details

ChatMol fine-tunes Llama3 backbones ranging from 1B to 8B parameters using LoRA, training on molecules drawn from ZINC250k and MOSES with prompts constructed for each constraint setting. Across single-property, substructure-property, and multi-property benchmarks it outperforms variational autoencoder and reinforcement learning baselines. In a multi-objective binding-affinity task against the estrogen receptor ESR1, ChatMol reaches a dissociation constant of 0.25 while retaining the best overall performance, a 4.76% improvement over prior methods. The numerical encoding contributes a Pearson correlation gain of up to 0.49 between instructed property values and the properties of generated molecules, the paper's clearest evidence that the mechanism does what it claims. Performance improves with backbone size, consistent with LLM scaling behavior. The authors state that code, data, and model weights will accompany the final version; as of the preprint no checkpoints or repository have been released.

#Applications

The natural fit is early-stage hit generation and lead optimization, where a medicinal chemist can state requirements in the terms they already use — retain this scaffold, push logP down, keep QED high — and receive candidate structures without configuring a separate optimizer for each objective. The substructure conditioning is what distinguishes it from latent-space samplers in practice: scaffold hopping and R-group enumeration around a fixed core map directly onto the prompt. The ESR1 case study shows the same interface driving a multi-objective affinity campaign against a named protein target.

#Impact

ChatMol belongs to a broader shift in which general-purpose language models, adapted cheaply, replace bespoke generative architectures for chemistry. Its specific contribution is diagnostic as much as generative: by isolating numerical representation as the failure mode and fixing it with an encoding change, the work quantifies how much of an LLM's difficulty with quantitative design instructions is representational rather than a limit of scale. The practical caveat is availability — until the promised code and weights appear, the results cannot be reproduced or built on, and evaluation to date is entirely in silico against computed property scores rather than measured assays.

At a glance

Released
February 2025
Category
Small molecule
Organization
Soochow University

Links

Research Paper

Tags

de_novo_designdrug_discoverygenerativelanguage_modeltransformer

Something wrong?

Much of this page is generated or calculated automatically. Flag anything that looks off and we will re-run it.