SoluProtMut: Siamese Deep Learning for Predicting Solubility Effects of Protein Mutations with Experimental Validation
Authors:
- Jan VeleckyJVJan Velecky
- Hana FaldynovaHFHana Faldynova
- Pedro HermosillaPHPedro Hermosilla
- Katerina SlanskaKSKaterina Slanska
- Mark DoerrMDMark Doerr
- Sara EgersdorfovaSESara Egersdorfova
- Uwe BornscheuerUwe Bornscheuer
- Jiří DamborskyJDJiří Damborsky
- Zbynek ProkopZPZbynek Prokop
- Stanislav MazurenkoSMStanislav Mazurenko
Preprint
Abstract
Protein solubility is an attractive engineering target because it is a critical property influencing the scalability of protein production and the success of therapeutic proteins in biomedical applications. However, predicting in silico solubility changes upon mutation is challenging due to data heterogeneity and protein bias. Here, we explore how different sources of solubility data can be used for machine learning and present SoluProtMut, a Siamese deep geometric neural network trained to predict the impact of mutations on protein solubility. Our final model was trained exclusively on deep mutational scanning data derived from recombinant protein expression assays in bacterial and yeast systems. We compare our model with five established solubility prediction methods. On an independent set of 44 proteins, the model achieves the highest informedness among the methods compared, with the largest margin for multipoint mutations (a ternary informedness of 26.5%, 8.8 percentage points above the next-best method). Our findings also reaffirm that the scarcity of solubility data continues to hamper progress in this field. To address this limitation, we experimentally quantified solubility changes for 277 single-point and multipoint mutants of haloalkane dehalogenase. Complemented by recent deep-mutational-scanning data on myoglobin, we employed both datasets for external validation. We identify protein-level generalization, rather than model capacity or training-set size, as the main limiting factor in the domain. Integrating high-throughput assays with deep learning nonetheless offers a route to improving the accuracy and scope of solubility prediction. Abstract Figure:Keywords
Funding Information
- Ministerstvo Školství, Mládeže a Tělovýchovy (CZ.02.01.01/00/23_029/0008437)
- Ministerstvo Školství, Mládeže a Tělovýchovy (LM2023069)
- Ministerstvo Školství, Mládeže a Tělovýchovy (LM2023055)
- Ministerstvo Školství, Mládeže a Tělovýchovy (LM2018140)
- Ministerstvo Školství, Mládeže a Tělovýchovy (LX22NPO5102)
- Grantová Agentura České Republiky (GX25-17329X)
- European Union’s Horizon 2020 Research and Innovation Programme (857560)
- European Union Centre of Excellence CLARA (101136607)
- Deutsche Forschungsgemeinschaft (DFG, INST 292/118-1 FUGG)
- The Federal State Mecklenburg-Vorpommern
This publication has 40 references indexed in Scilit:
- Deep mutational scanning reveals a de novo disulfide bond and combinatorial mutations for engineering thermostable myoglobinProtein Science, 2025
- Predicting the effects of mutations on protein solubility using graph convolution network and protein language model representationJournal of Computational Chemistry, 2023
- Siamese Neural Networks: An Overview2020
- Fully automatized high‐throughput enzyme library screening using a robotic platformBiotechnology and Bioengineering, 2016
- Error-Prone PCR and Effective Generation of Gene Variant Libraries for Directed Evolution2014
- ggplot2WIREs Computational Statistics, 2011
- Recent Advances in GFP Folding Reporter and Split-GFP Solubility Reporter Technologies. Application to Improving the Folding and Solubility of Recalcitrant Proteins from Mycobacterium tuberculosisJournal of Structural and Functional Genomics, 2005
- A new set of amino acid descriptors and its application in peptide QSARsPeptide Science, 2005
- Challenges in the development of high protein concentration formulationsJournal of Pharmaceutical Sciences, 2004
- New fusion protein systems designed to give soluble expression inEscherichia coliBiotechnology and Bioengineering, 1999