⋅The halophyte Suaeda maritima is well-known for its potent antioxidant and antimicrobial properties, typically validated through crude extracts. However, identifying the specific protein-derived peptides responsible for this efficacy remains a challenge. This study establishes a high-throughput pipeline starting with transcriptome sequencing to predict the peptide function of S. maritima. These sequences were computationally fragmented into a peptide library, providing a molecular foundation for bioactive discovery.

The core of our approach involved calculating high-dimensional biochemical features for each peptide. We then deployed machine learning (ML) models trained to recognize functional patterns within these descriptors, enabling high-precision prediction of bioactivity scores. To ensure biological relevance, we integrated pan-genome analysis to identify conserved core sequences and DEG (Differentially Expressed Gene) data to prioritize peptides up-regulated under environmental stress.

This integrated workflow successfully narrowed down thousands of sequences to a few high-confidence candidates with defined molecular structures. By combining AI-driven prediction with genomic validation, this study transforms the known efficacy of S. maritimainto precise, sequence-defined materials. These findings offer promising candidates for next-generation functional additives in the pharmaceutical and cosmeceutical industries.

⋅Funding: This work was supported by a grant from the Honam National Institute of Biological Resources (HNIBR), funded by the Ministry of Climate, Energy and Environment (MCEE), Republic of Korea (HNIBR2026-B-1-15), and by the Korea Environment Industry & Technology Institute (KEITI) through the project Advancement of Multi-ministerial National Biological Research Resources, funded by the Ministry of Climate, Energy and Environment (MCEE) (RS-2023-00230404).