TastepepAI: An artificial intelligence platform for taste peptide de novo design

Jianda Yue, Tingting Li, Jian Ouyang, Jiawei Xu, Hua Tan, Zihui Chen, Changsheng Han, Huanyu Li, Songping Liang, Zhonghua Liu, Zhonghua Liu and Ying Wang

PLOS Computational Biology, 2025, vol. 21, issue 10, 1-23

Abstract: Taste peptides have emerged as promising natural flavoring agents attributed to their unique organoleptic properties, high safety profile, and potential health benefits. However, the de novo identification of taste peptides derived from animal, plant, or microbial sources remains a time-consuming and resource-intensive process, significantly impeding their widespread application in the food industry. In this work, we present TastePepAI, a comprehensive artificial intelligence framework for customized taste peptide design and safety assessment. As the key element of this framework, a loss-supervised adaptive variational autoencoder (LA-VAE) is implemented to efficiently optimize the latent representation of sequences during training and facilitate the generation of target peptides with desired taste profiles. Notably, our model incorporates a novel taste-avoidance mechanism, allowing for selective flavor exclusion. Subsequently, our in-house developed toxicity prediction algorithm (SpepToxPred) is integrated in the framework to undergo rigorous safety evaluation of generated peptides. Using this integrated platform, we successfully identified 73 peptides exhibiting sweet, salty, and umami, significantly expanding the current repertoire of taste peptides. This work demonstrates the potential of TastePepAI in accelerating taste peptide discovery for food applications and provides a versatile framework adaptable to broader peptide engineering challenges.Author summary: Taste peptides have established themselves as attractive natural flavor enhancers, thanks to their distinct sensory attributes, strong safety record, and possible health advantages. TastePepAI, the first artificial intelligence platform for designing taste peptides with desired flavor profiles, was developed in this work. Traditional methods for identifying taste peptides are time-consuming and costly, with their applications in the food industry limited. Two key innovations are featured in our integrated computational framework: LA-VAE, which is used for generating peptide sequences with target taste properties while suppressing unwanted characteristics, and SpepToxPred for safety assessment—with its accuracy being 12% higher than that of existing toxicity prediction models. Using this platform, 73 novel multifunctional taste peptides exhibiting sweet, salty, and umami properties were successfully designed and validated. Electronic tongue analysis confirmed their expected taste characteristics, while safety assays demonstrated excellent biocompatibility. To promote open science, we established the TastePepMap database and TastePepAI design platform. This work demonstrates AI’s potential in functional peptide design and provides crucial methodological foundations for developing next-generation peptide-based taste modulators, offering new opportunities for creating healthier and more sustainable food ingredients.

Date: 2025
References: Add references at CitEc
Citations:

Downloads: (external link)
https://journals.plos.org/ploscompbiol/article?id=10.1371/journal.pcbi.1013602 (text/html)
https://journals.plos.org/ploscompbiol/article/fil ... 13602&type=printable (application/pdf)

Related works:
This item may be available elsewhere in EconPapers: Search for items with the same title.

Export reference: BibTeX RIS (EndNote, ProCite, RefMan) HTML/Text

Persistent link: https://EconPapers.repec.org/RePEc:plo:pcbi00:1013602

DOI: 10.1371/journal.pcbi.1013602

Access Statistics for this article

More articles in PLOS Computational Biology from Public Library of Science
Bibliographic data for series maintained by ploscompbiol ().