EUVIMEDEUVIMED entdecken
Open Access · CC BY 4.0

LoMuS: low-rank adaptation with sequence multi-representation improves protein stability prediction

Samuel Infante, Akash Singh, Anowarul Kabir iD

Bioinformatics · 2026

DOI: 10.1093/bioinformatics/btag509 ↗

Kurz gefasst

Fachartikel in der Zeitschrift Bioinformatics, 2026. Verfasst von Samuel Infante und 2 weiteren Autor:innen. Beteiligte Einrichtung: Bellini College of AI, Cybersecurity and Computing, University of South Florida, Tampa, Florida 33620, United States. Hauptthemen laut MEDLINE: Proteins, Computational Biology, Deep Learning, Sequence Analysis, Protein. Die Arbeit wurde bisher 1-mal zitiert (Crossref) und nennt 53 Literaturangaben. Der Volltext ist frei zugänglich (Open Access, CC BY 4.0). Gefördert von: University of South Florida, FL, USA.

AbstractSchlagwörterBibliografische AngabenZitierenVerwandte Studien

Abstract (Originalsprache)

Motivation Protein folding stability is a key determinant for understanding protein dynamics, including molecular function, pathogenicity, and protein engineering. Yet, accurate prediction of protein stability remains challenging due to high variability in available data, particularly when only sequence information is available and structural knowledge is limited or unavailable. In this work, we introduce LoMuS, a multi-representation-based deep learning model that predicts dataset-provided protein stability scores directly from the primary sequence. In the core of the model architecture, a fusion network integrates explicit physicochemical descriptors with low-rank adapted protein language model derived embeddings from the sequence that consistently gains across standard experimental stability benchmarks. Results We rigorously evaluate LoMuS across multiple settings, such as absolute folding stability scoring, mutation landscape stability scoring, held-out protein domains, out-of-distribution label regimes, and per-protein evaluation. LoMuS consistently outperforms sequence-only baselines, achieving an absolute performance gain of at least 10% in Spearman’s rank correlation across several benchmarks. Per-protein evaluations further demonstrate robust performance gains. Ablation analyses confirm that complementary signals from physicochemical descriptors and sequence embeddings are critical to the effectiveness of the proposed multi-representation approach. We believe LoMuS advances protein engineering research by improving the prediction and ranking of protein stability scores. Availability All codes including data preparation scripts, training and validation recipes, and experimental configurations for LoMuS are available at: https://github.com/kabir-ai2bio-lab/LoMuS.

Schlagwörter

MeSH-Schlagwörter (MEDLINE)

ProteinsComputational BiologyDeep LearningSequence Analysis, ProteinProtein StabilityProtein FoldingPrediction AlgorithmsAmino Acid Sequence

Substanzen

Proteins

Autor:innen und Institutionen

  • Bellini College of AI, Cybersecurity and Computing, University of South Florida, Tampa, Florida 33620, United States.

ORCID-Profile: Anowarul Kabir

Förderung

  • University of South Florida, FL, USA

Zitieren

APA

Infante, S., Singh, A., & Kabir, A. (2026). LoMuS: low-rank adaptation with sequence multi-representation improves protein stability prediction. Bioinformatics. https://doi.org/10.1093/bioinformatics/btag509

Vancouver

Infante S, Singh A, Kabir A. LoMuS: low-rank adaptation with sequence multi-representation improves protein stability prediction. Bioinformatics. 2026. doi:10.1093/bioinformatics/btag509

BibTeX
@article{Infante2026,
  author = {Infante, Samuel and Singh, Akash and Kabir, Anowarul},
  title = {LoMuS: low-rank adaptation with sequence multi-representation improves protein stability prediction},
  journal = {Bioinformatics},
  year = {2026},
  doi = {10.1093/bioinformatics/btag509}
}
Kostenlos starten →