# Representing Molecules with Algebraic Data Types: Beyond SMILES and SELFIES

arXiv: https://arxiv.org/abs/2501.13633 · HTML: https://mlsanity.com/paper/2501.13633/

## Key facts

- No citing papers in the map yet (data through 14 Oct 2026).
- Builds on 16 of its 94 references that are in the map.
- First posted on arXiv 23 Jan 2025.

## Builds on (most cited first, 16 of 16)

- [What Uncertainties Do We Need in Bayesian Deep Learning for Computer Vision?](https://mlsanity.com/paper/1703.04977/) (2017)
- [Geometric Deep Learning: Grids, Groups, Graphs, Geodesics, and Gauges](https://mlsanity.com/paper/2104.13478/) (2021)
- [Automatic chemical design using a data-driven continuous representation of molecules](https://mlsanity.com/paper/1610.02415/) (2016)
- [GeoDiff: a Geometric Diffusion Model for Molecular Conformation Generation](https://mlsanity.com/paper/2203.02923/) (2022)
- [Pyro: Deep Universal Probabilistic Programming](https://mlsanity.com/paper/1810.09538/) (2018)
- [ChemRL-GEM: Geometry Enhanced Molecular Representation Learning for Property Prediction](https://mlsanity.com/paper/2106.06130/) (2021)
- [Learning to Plan Chemical Syntheses](https://mlsanity.com/paper/1708.04202/) (2017)
- [SELFIES and the future of molecular string representations](https://mlsanity.com/paper/2204.00056/) (2022)
- [An Introduction to Probabilistic Programming](https://mlsanity.com/paper/1809.10756/) (2018)
- [Molecular Hypergraph Grammar with its Application to Molecular Optimization](https://mlsanity.com/paper/1809.02745/) (2018)
- [Artificial Intelligence in Drug Discovery: Applications and Techniques](https://mlsanity.com/paper/2106.05386/) (2021)
- [Simple, Distributed, and Accelerated Probabilistic Programming](https://mlsanity.com/paper/1811.02091/) (2018)
- [Gaussian Process Molecule Property Prediction with FlowMO](https://mlsanity.com/paper/2010.01118/) (2020)
- [CHAD: Combinatory Homomorphic Automatic Differentiation](https://mlsanity.com/paper/2103.15776/) (2021)
- [Modular Probabilistic Models via Algebraic Effects](https://mlsanity.com/paper/2203.04608/) (2022)
- [Modular probabilistic programming with algebraic effects (MSc Thesis 2019)](https://mlsanity.com/paper/2412.19826/) (2024)

## Abstract (arXiv)

Benchmarks of molecular machine learning models often treat the molecular representation as a neutral input format, yet the representation defines the syntax of validity, edit operations, and invariances that models implicitly learn. We propose MolADT, a typed intermediate representation (IR) for molecules expressed as a family of algebraic data types that separates (i) constitution via Dietz-style bonding systems, (ii) 3D geometry and stereochemistry, and (iii) optional electronic annotations. By shifting from string edits to operations over structured values, MolADT makes representational assumptions explicit, supports deterministic validation and localized transformations, and provides hooks for symmetry-aware and Bayesian workflows. We provide a reference implementation in Haskell (open-source, archived with DOI) and worked examples demonstrating delocalised/multicentre bonding, validation invariants, reaction extensions, and group actions relevant to geometric learning.