Anti-hypertensive Peptide Predictor: A Machine Learning-Empowered Web Server for Prediction of Food-Derived Peptides with Potential Angiotensin-Converting Enzyme-I Inhibitory Activity

Abstract

Angiotensin converting enzyme-I (ACE-I) is a key therapeutic target of the renin−angiotensin−aldosterone system (RAAS), the central pathway of blood pressure regulation. Food-derived peptides with ACE-I inhibitory activities are receiving significant research attention. However, identification of ACE-I inhibitory peptides from different food proteins is a labor-intensive, lengthy, and expensive process. For successful identification of potential ACE-I inhibitory peptides from food sources, a machine learning and structural bioinformatics-based web server has been developed and reported in this study. The web server can take input in the FASTA format or through UniProt ID to perform the in silico gastrointestinal digestion and then screen the resulting peptides for ACE-I inhibitory activity. This unique platform provides elaborated structural and functional features of the active peptides and their interaction with ACE-I. Thus, it can potentially enhance the efficacy and reduce the time and cost in identifying and characterizing novel ACE-I inhibitory peptides from food proteins. URL: http://hazralab.iitr.ac.in/ahpp/index.php.

Isolation, identification and in vivo antihypertensive effect of novel angiotensin I-converting enzyme (ACE) inhibitory peptides from Spirulina protein hydrolysate

“…33 Recently, Gazal Kalyan et al (2021) summarized the reported ACE inhibitory peptides and their IC 50 values. 34 Out of the 1648 ACE inhibitory peptides, only 117 peptides potently inhibited ACE with an IC 50 value below 3 μM, and about 43% peptides had an IC 50 lower than 60 μM. 34 The novelty and potency of the oligopeptides SpH-6 and SpH-7 revealed in this study demonstrated the efficiency by applying combined virtual screening and in vitro experimental investigation for the discovery of bioactive peptides from various dietary protein resources.…”

MLACP 2.0: An updated machine learning tool for anticancer peptide prediction

“…Several factors contribute to the improved performance of MLACP 2.0, including (i) a reduced training dataset coupled with a meta -model approach; (ii) the predicted probabilistic features have a high intrinsic discriminatory ability on both datasets, resulting in improved performance. Interestingly, this approach can be extended to predict other peptide therapeutic functions [51] , [52] , [53] , [54] . Despite its promising performance, MLACP 2.0 also has room for improvement.…”

Comprehensive Evaluation and Comparison of Machine Learning Methods in QSAR Modeling of Antioxidant Tripeptides

“… 40 , 41 LOOCV was used to avoid overfitting and tune the hyperparameters because of our small data set size. 8 , 14 , 33 The hyperparameters with the best performance from LOOCV were used as the final model for performance evaluation with the test data set.…”