EUVIMEDOpdag EUVIMED
Open access · CC BY 4.0

Persistent sheaf Laplacian analysis of protein stability and solubility changes upon mutation

Yiming Ren, Junjie Wee iD, Xi Chen, Grace Qian, Guo-Wei Wei iD

Protein Science · 2026-08 · 35 (8)

DOI: 10.1002/pro.70700 ↗

At a glance

Journal article in Protein Science, published online on 2026-07-10. Written by Yiming Ren and 4 co-authors. Institution: Department of Mathematics, Michigan State University, East Lansing, Michigan, USA. Major MEDLINE topics: Proteins, Mutation, Deep Learning. Cited 1 times so far (Crossref); 90 references. The full text is freely available (open access, CC BY 4.0). Funded by: Georgia Research Alliance, NIH HHS, NIGMS NIH HHS.

AbstractSubjectsBibliografiske oplysningerCiteRelated studies

Abstract (original language)

Genetic mutations frequently disrupt protein structure, stability, and solubility, acting as primary drivers for a wide spectrum of diseases. Despite the critical importance of these molecular alterations, existing computational models often lack interpretability and fail to integrate essential physicochemical interactions. To overcome these limitations, we propose SheafLapNet, a predictive framework grounded in the mathematical theory of Topological Deep Learning (TDL) and Persistent Sheaf Laplacian (PSL). Unlike standard Topological Data Analysis (TDA) tools such as persistent homology, which are often insensitive to heterogeneous information, PSL explicitly encodes specific physical and chemical information such as partial charges directly into the topological analysis. SheafLapNet synergizes these sheaf‐theoretic invariants with advanced protein transformer features and auxiliary physical descriptors to capture intrinsic molecular interactions in a multiscale and mechanistic manner. To validate our framework, we employ rigorous benchmarks for both regression and classification tasks. For stability prediction, we utilize the comprehensive S2648 dataset, alongside the independent S350 and strictly non‐redundant S669 blind test sets to ensure robust evaluation and thermodynamic consistency. For solubility prediction, we employ the PON‐Sol2 dataset, which provides annotations for increased, decreased, or neutral solubility changes. By integrating these multi‐perspective features, SheafLapNet achieves state‐of‐the‐art performance across these diverse benchmarks, demonstrating that sheaf‐theoretic modeling significantly enhances both interpretability and generalizability in predicting mutation‐induced structural and functional changes.

Subjects

MeSH terms (MEDLINE)

ProteinsMutationDeep LearningSolubilityProtein StabilityThermodynamics

Author keywords

mutationpersistent topological Laplaciansprotein folding stabilityprotein solubilitysheaf Laplacian networks

Substances

Proteins

Authors and institutions

  • Department of Mathematics, Michigan State University, East Lansing, Michigan, USA.
  • The Frazer School, Gainesville, Florida, USA.
  • Lassiter High School, Marietta, Georgia, USA.
  • Department of Mathematics, University of Georgia, Athens, Georgia, USA.
  • Department of Biochemistry and Molecular Biology, University of Georgia, Athens, Georgia, USA.

ORCID profiles: Junjie Wee, Guo-Wei Wei

Funding

  • Georgia Research Alliance
  • NIH HHS
  • NIGMS NIH HHS
  • Michigan State University Research Foundation
  • University of Georgia
  • NIAID NIH HHS
  • National Institutes of Health

Cite

APA

Ren, Y., Wee, J., Chen, X., Qian, G., & Wei, G. W. (2026). Persistent sheaf Laplacian analysis of protein stability and solubility changes upon mutation. Protein Science, 35(8). https://doi.org/10.1002/pro.70700

Vancouver

Ren Y, Wee J, Chen X, Qian G, Wei GW. Persistent sheaf Laplacian analysis of protein stability and solubility changes upon mutation. Protein Science. 2026;35(8). doi:10.1002/pro.70700

BibTeX
@article{Ren2026,
  author = {Ren, Yiming and Wee, Junjie and Chen, Xi and Qian, Grace and Wei, Guo-Wei},
  title = {Persistent sheaf Laplacian analysis of protein stability and solubility changes upon mutation},
  journal = {Protein Science},
  year = {2026},
  volume = {35},
  number = {8},
  doi = {10.1002/pro.70700}
}
Start for free →