Machine learning can guide experimental approaches for protein digestibility estimations

TL;DR

This paper introduces a machine learning model that predicts food protein digestibility with 90% accuracy, potentially replacing lengthy experimental methods and accelerating the development of sustainable, nutritious food sources.

Contribution

It presents the first AI-based model combining nutritional data, protein sequences, and interpretability techniques to accurately predict protein digestibility.

Findings

Achieves 90% accuracy in digestibility prediction

Uses a novel combination of biochemical properties and protein embeddings

Reduces reliance on in-vivo and in-vitro experiments

Abstract

Food protein digestibility and bioavailability are critical aspects in addressing human nutritional demands, particularly when seeking sustainable alternatives to animal-based proteins. In this study, we propose a machine learning approach to predict the true ileal digestibility coefficient of food items. The model makes use of a unique curated dataset that combines nutritional information from different foods with FASTA sequences of some of their protein families. We extracted the biochemical properties of the proteins and combined these properties with embeddings from a Transformer-based protein Language Model (pLM). In addition, we used SHAP to identify features that contribute most to the model prediction and provide interpretability. This first AI-based model for predicting food protein digestibility has an accuracy of 90% compared to existing experimental techniques. With this accuracy, our model can eliminate the need for lengthy in-vivo or in-vitro experiments, making the process of creating new foods faster, cheaper, and more ethical.