AnOxPePred: using deep learning for the prediction of antioxidative properties of peptides

Journal article

Five native rice varieties from Thailand were used to extract seed storage proteins. The most prevalent proteins were glutelin, globulin, albumin, and prolamin. Significant scavenging activity of crude glutelin was seen in all samples, and this activity was further improved in hydrolysates made with pepsin. Using PeptideCutter (https://web.expasy.org/peptide_cutter/), 99 peptides were produced by simulating the gastrointestinal digestion of rice glutelin (Oryza sativa Indica Group; GenBank: AGT59178.1). AnOxPePred-1.0 (https://services.healthtech.dtu.dk/services/AnOxPePred-1.0/) was used to predict the antioxidant potential of many peptides. The top five peptides with high ABTS•+ and DPPH• scavenging activities were examined for molecular docking. The results indicate that, when compared to glutathione, the positive control, the octapeptide TNTPGVVY had the lowest binding affinity with DPPH• (-4.26 kcal/mol), and the 13-amino acid peptide, TQQQEQAQAQDQY, had the lowest binding affinity with ABTS•+ (-3.70 kcal/mol). The antioxidant properties of both synthetic peptides were confirmed by in vitro assays. Neither peptide exhibited cytotoxic effects on human cell lines. This research indicates the value of Thai red glutinous rice and its potential to be developed into health food products for those who love eating sticky rice

Authors: Olsen, TH; Yesiltas, B; Marin, FI; Pertseva, M; García-Moreno, PJ

• Publication status: Published
• Peer review status: Peer reviewed
• Publisher: Nature Research
• Journal: Scientific Reports
• Volume: 10
• Issue: 1
• Pages: 21471-21471
• Publication date: 2020-12-08
• DOI: 10.1038/s41598-020-78319-w
• EISSN: 2045-2322
• ISSN: 2045-2322
• Language: English
• Keywords: Computer science; Chemistry; Artificial intelligence