CaLMPhosKAN: prediction of general phosphorylation sites in proteins via fusion of codon aware embeddings with amino acid aware embeddings and wavelet-based Kolmogorov–Arnold network
Bioinformatics · 2025
DOI: 10.1093/bioinformatics/btaf124 ↗Kurz gefasst
Fachartikel in der Zeitschrift Bioinformatics, 2025. Verfasst von Pawel Pratyush und 5 weiteren Autor:innen. Beteiligte Einrichtung: Golisano College of Computing and Information Sciences, Rochester Institute of Technology, Rochester, NY 14623, United States. Hauptthemen laut MEDLINE: Codon, Proteins, Amino Acids, Computational Biology. Die Arbeit wurde bisher 3-mal zitiert (Crossref) und nennt 36 Literaturangaben. Der Volltext ist frei zugänglich (Open Access, CC BY 4.0). Gefördert von: National Science Foundation.
Abstract (Originalsprache)
Motivation The mapping from codon to amino acid is surjective due to codon degeneracy, suggesting that codon space might harbor higher information content. Embeddings from the codon language model have recently demonstrated success in various protein downstream tasks. However, predictive models for residue-level tasks such as phosphorylation sites, arguably the most studied Post-Translational Modification (PTM), and PTM sites prediction in general, have predominantly relied on representations in amino acid space. Results We introduce a novel approach for predicting phosphorylation sites by utilizing codon-level information through embeddings from the codon adaptation language model (CaLM), trained on protein-coding DNA sequences. Protein sequences are first reverse-translated into reliable coding sequences by mapping UniProt sequences to their corresponding NCBI reference sequences and extracting the exact coding sequences from their GenBank format using a dynamic programming-based global pairwise alignment. The resulting coding sequences are encoded using the CaLM encoder to generate codon-aware embeddings, which are subsequently integrated with amino acid-aware embeddings obtained from a protein language model, through an early fusion strategy. Next, a window-level representation of the site of interest, retaining the full sequence context, is constructed from the fused embeddings. A ConvBiGRU network extracts feature maps that capture spatiotemporal correlations between proximal residues within the window. This is followed by a prediction head based on a Kolmogorov-Arnold network (KAN) using the derivative of gaussian wavelet transform to generate the inference for the site. The overall model, dubbed CaLMPhosKAN, performs better than the existing approaches across multiple datasets. Availability and implementation CaLMPhosKAN is publicly available at https://github.com/KCLabMTU/CaLMPhosKAN.
Schlagwörter
MeSH-Schlagwörter (MEDLINE)
Substanzen
Autor:innen und Institutionen
- Golisano College of Computing and Information Sciences, Rochester Institute of Technology, Rochester, NY 14623, United States.
- College of Computing, Michigan Technological University, Houghton, MI 49931, United States.
- College of Engineering, North Carolina Agricultural and Technical State University, Greensboro, NC 27411, United States.
- College of Applied Sciences and Technology, Illinois State University, Normal, IL 61761, United States.
ORCID-Profile: Pawel Pratyush, Suresh Pokharel, Hamid D Ismail, Meenal Chaudhari, Dukka B Kc
Förderung
- National Science Foundation
Zitieren
APA
Pratyush, P., Carrier, C., Pokharel, S., Ismail, H. D., Chaudhari, M., & Kc, D. B. (2025). CaLMPhosKAN: prediction of general phosphorylation sites in proteins via fusion of codon aware embeddings with amino acid aware embeddings and wavelet-based Kolmogorov–Arnold network. Bioinformatics. https://doi.org/10.1093/bioinformatics/btaf124
Vancouver
Pratyush P, Carrier C, Pokharel S, Ismail HD, Chaudhari M, Kc DB. CaLMPhosKAN: prediction of general phosphorylation sites in proteins via fusion of codon aware embeddings with amino acid aware embeddings and wavelet-based Kolmogorov–Arnold network. Bioinformatics. 2025. doi:10.1093/bioinformatics/btaf124
BibTeX
@article{Pratyush2025,
author = {Pratyush, Pawel and Carrier, Callen and Pokharel, Suresh and Ismail, Hamid D and Chaudhari, Meenal and Kc, Dukka B},
title = {CaLMPhosKAN: prediction of general phosphorylation sites in proteins via fusion of codon aware embeddings with amino acid aware embeddings and wavelet-based Kolmogorov–Arnold network},
journal = {Bioinformatics},
year = {2025},
doi = {10.1093/bioinformatics/btaf124}
}
EUVIMED