Intestinal microbiota alterations by dietary exposure to chemicals from food cooking and processing. Application of data science for risk prediction

Abstract

Diet is one of the main sources of exposure to toxic chemicals with carcinogenic potential, some of which are generated during food processing, depending on the type of food (primarily meat, fish, bread and potatoes), cooking methods and temperature. Although demonstrated in animal models at high doses, an unequivocal link between dietary exposure to these compounds with disease has not been proven in humans. A major difficulty in assessing the actual intake of these toxic compounds is the lack of standardised and harmonised protocols for collecting and analysing dietary information. The intestinal microbiota (IM) has a great influence on health and is altered in some diseases such as colorectal cancer (CRC). Diet influences the composition and activity of the IM, and the net exposure to genotoxicity of potential dietary carcinogens in the gut depends on the interaction among these compounds, IM and diet. This review analyses critically the difficulties and challenges in the study of interactions among these three actors on the onset of CRC. Machine Learning (ML) of data obtained in subclinical and precancerous stages would help to establish risk thresholds for the intake of toxic compounds generated during food processing as related to diet and IM profiles, whereas Semantic Web could improve data accessibility and usability from different studies, as well as helping to elucidate novel interactions among those chemicals, IM and diet.

Keywords: Colorectal cancer; Diet; Intestinal microbiota; Machine learning; Semantic web; Toxic chemicals.

Fig. 1

Schematic representation of risk assessment by exposure to dietary toxic compounds formed during food cooking and processing as a function of the IM, diet and intestinal toxicity, applying ML and Semantic Web. The net exposure to toxic compounds depends on the intake and time of exposure and this influences the genotoxicity at the intestinal environment. IM and global diet could modify the resulting toxicity of dietary chemicals. Prolonged exposure to high intestinal toxicity levels could lead to changes in the intestinal mucosa that may be accompanied by shifts in the intestinal microbiota. Applying ML to dietary and microbiota data in silent, subclinical and precancerous stages of intestinal mucosal damage could assist in CRC risk assessment whereas Semantic Web will facilitate data accessibility and management.

Fig. 2

General workflow of a Machine Learning process for CRC risk assessment as a function of diet, microbiota and intestinal genotoxicity. Data from diet (FFQ), microbial metabolites, microbiota composition, microbial gene functions, and genotoxicity/mutagenicity (faeces) and biopsia analyses of the intestinal mucosa (routine colonoscopies at hospitals) are collected in a joint database and submitted to a ML process. Some ML models (such as DT, on bottom-left) allow establishing profiles and thresholds related to the input variables, while others (such as ANN, on bottom-right) are more difficult to interpret but are successful predictors.

Fig. 3

Semantic Web schema and technological stack proposed for microbiota and diet studies. Each concentric circumference represents a layer/process in the technological stack; these layers are independent and can work by themselves. The layer stacking means that an upper layer contains the lower ones and need for them to be complete and coherent. Different coloured graphs represent graphs from different sources, which are not yet integrated. Orange and yellow patterns in the validation phase represent the mechanism of validation and normalization of the aforementioned heterogeneous graphs, which connect to a unique and integrated knowledge graph. (For interpretation of the references to colour in this figure legend, the reader is referred to the web version of this article.)