# Compressing Chemistry Reveals Functional Groups

arXiv: https://arxiv.org/abs/2511.05728 · HTML: https://mlsanity.com/paper/2511.05728/

## Key facts

- No citing papers in the map yet (data through 14 Oct 2026).
- Builds on 7 of its 57 references that are in the map.
- First posted on arXiv 7 Nov 2025.

## Builds on (most cited first, 7 of 7)

- [Convolutional Networks on Graphs for Learning Molecular Fingerprints](https://mlsanity.com/paper/1509.09292/) (2015)
- [Large-Scale Chemical Language Representations Capture Molecular Structure and Properties](https://mlsanity.com/paper/2106.09553/) (2021)
- [Learning programs by learning from failures](https://mlsanity.com/paper/2005.02259/) (2020)
- [Learning MDL logic programs from noisy data](https://mlsanity.com/paper/2308.09393/) (2023)
- [Meta-QSAR: a large-scale application of meta-learning to drug design and discovery](https://mlsanity.com/paper/1709.03854/) (2017)
- [Deep Confidence: A Computationally Efficient Framework for Calculating Reliable Errors for Deep Neural Networks](https://mlsanity.com/paper/1809.09060/) (2018)
- [Learning Logical Rules using Minimum Message Length](https://mlsanity.com/paper/2508.06230/) (2025)

## Abstract (arXiv)

We introduce the first formal large-scale assessment of the utility of traditional chemical functional groups as used in chemical explanations. Our assessment employs a fundamental principle from computational learning theory: a good explanation of data should also compress the data. We introduce an unsupervised learning algorithm based on the Minimum Message Length (MML) principle that searches for substructures that compress around three million biologically relevant molecules. We demonstrate that the discovered substructures contain most human-curated functional groups as well as novel larger patterns with more specific functions. We also run our algorithm on 24 specific bioactivity prediction datasets to discover dataset-specific functional groups. Fingerprints constructed from dataset-specific functional groups are shown to significantly outperform other fingerprint representations, including the MACCS and Morgan fingerprint, when training ridge regression models on bioactivity regression tasks.