# A Pipeline for Data-Driven Learning of Topological Features with Applications to Protein Stability Prediction

arXiv: https://arxiv.org/abs/2408.04847 · HTML: https://mlsanity.com/paper/2408.04847/

## Key facts

- Cited by 1 papers in the map (data through 14 Oct 2026).
- Builds on 12 of its 45 references that are in the map.
- First posted on arXiv 9 Aug 2024.

## Builds on (most cited first, 12 of 12)

- [Persistence Images: A Stable Vector Representation of Persistent Homology](https://mlsanity.com/paper/1507.06217/) (2015)
- [A roadmap for the computation of persistent homology](https://mlsanity.com/paper/1506.08903/) (2015)
- [A Stable Multi-Scale Kernel for Topological Machine Learning](https://mlsanity.com/paper/1412.6821/) (2014)
- [Correlation and variable importance in random forests](https://mlsanity.com/paper/1310.5726/) (2013)
- [A persistent homology approach to heart rate variability analysis with an application to sleep-wake classification](https://mlsanity.com/paper/1908.06856/) (2019)
- [Approximating Continuous Functions on Persistence Diagrams Using Template Functions](https://mlsanity.com/paper/1902.07190/) (2019)
- [PersGNN: Applying Topological Data Analysis and Geometric Deep Learning to Structure-Based Protein Function Prediction](https://mlsanity.com/paper/2010.16027/) (2020)
- [Minimal Cycle Representatives in Persistent Homology using Linear Programming: an Empirical Study with User's Guide](https://mlsanity.com/paper/2105.07025/) (2021)
- [Stability and Machine Learning Applications of Persistent Homology Using the Delaunay-Rips Complex](https://mlsanity.com/paper/2303.01501/) (2023)
- [Adaptive template systems: Data-driven feature selection for learning with persistence diagrams](https://mlsanity.com/paper/1910.06741/) (2019)
- [Topological and Statistical Behavior Classifiers for Tracking Applications](https://mlsanity.com/paper/1406.0214/) (2014)
- [Supervised Learning of Labeled Pointcloud Differences via Cover-Tree Entropy Reduction](https://mlsanity.com/paper/1702.07959/) (2017)

## Built on by (most cited first, 1 of 1)

- [LandscapeSHAP: Which Persistent Homology Class Gets the Credit?](https://mlsanity.com/paper/2609.31469/) (2026)

## Abstract (arXiv)

In this paper, we propose a data-driven method to learn interpretable topological features of biomolecular data and demonstrate the efficacy of parsimonious models trained on topological features in predicting the stability of synthetic mini proteins. We compare models that leverage automatically-learned structural features against models trained on a large set of biophysical features determined by subject-matter experts (SME). Our models, based only on topological features of the protein structures, achieved 92%-99% of the performance of SME-based models in terms of the average precision score. By interrogating model performance and feature importance metrics, we extract numerous insights that uncover high correlations between topological features and SME features. We further showcase how combining topological features and SME features can lead to improved model performance over either feature set used in isolation, suggesting that, in some settings, topological features may provide new discriminating information not captured in existing SME features that are useful for protein stability prediction.