A systematic framework for identifying amphipathic cationic motifs in the human proteome reveals a distinct non-random physicochemical class

Read full text Summarise with VIRAT
Authors: Takehito Kojima
Cited by: 0
References: 9

Summary

No summary yet. VIRAT can read this paper and write a plain-language summary of its aim, methods, key findings and limitations.

Summarise with VIRAT Free to try with your Science Catalogs account.

Abstract

Abstract Short amphipathic helices are widely recognized as functional motifs involved in membrane interaction and antimicrobial activity. However, their global distribution and structural characteristics within the human proteome remain incompletely understood. In this study, we systematically analyzed the human proteome (~ 20,000 proteins) using a fixed sliding window of 12 residues to identify candidate amphipathic motifs. Each segment was evaluated based on hydrophobic moment, mean hydrophobicity, and net charge, followed by stringent physicochemical filtering. This approach reduced the search space to 475 candidate motifs exhibiting strong amphipathic properties, characterized by high positive charge ( ≥ + 6) and moderate hydrophobic moment (~ 0.45–0.55). Functional annotation revealed enrichment in membrane-associated proteins, while a substantial fraction of candidates remained unclassified, suggesting the presence of previously unrecognized functional elements. Positional analysis demonstrated a clear bias toward internal (middle) regions of proteins, supporting a role as structural or interaction modules rather than signal peptides. Comparison with randomly sampled peptide segments confirmed that the identified motifs occupy a distinct physicochemical space, indicating that their properties are not attributable to random sequence composition. These findings demonstrate that amphipathic cationic motifs with antimicrobial-like physicochemical signatures are widely embedded within human proteins. This systematic framework provides a basis for identifying cryptic functional elements and may inform the design of antimicrobial peptides and membrane-interacting biomolecules.

Fields of Study & Subjects

Antimicrobial Peptides and Activities Protein Structure and Dynamics Machine Learning in Bioinformatics

Publication details

Journal / source
Research Square
Publication date
25 March 2026
Language
EN
Document type
Preprint