# Flexible Kernels for Protein Property Prediction

arXiv: https://arxiv.org/abs/2606.11057 · HTML: https://mlsanity.com/paper/2606.11057/

## Key facts

- Cited by 1 papers in the map (data through 14 Oct 2026).
- Builds on 0 of its 79 references that are in the map.
- First posted on arXiv 9 Jun 2026.

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

- [Information-Dense Synthesis for Molecular Discovery](https://mlsanity.com/paper/2610.08495/) (2026)

## Abstract (arXiv)

Despite its importance to applications in protein design, predicting protein properties like binding affinity and thermostability from sparse experimental data remains a significant challenge. Accordingly, we introduce a class of sequence kernels that exploit evolutionary substitution matrices as well as local linearity and demonstrate that the resulting Gaussian processes provide data-efficient models of protein property landscapes, frequently outperforming alternatives that rely on foundation model embeddings. Furthermore--by learning what are in effect structure-aware substitution matrices--we show that our kernels can readily incorporate structural information from foundation models. We demonstrate that these structure-conditioned kernels are well suited to multi-task learning across multiple protein property landscapes and can decisively outperform local supervised learning methods.