DIYUFOOD

A healthy, balanced diet plays a crucial role in maintaining metabolic health and supporting the immune system. Obesity has been linked to various metabolic syndromes, such as dyslipidemia, hyperglycemia, and fatty liver. The presence of hyperglycemia and lipid accumulation can stimulate lipid oxidation, resulting in the overproduction of cytokines, hyperactivation of the complement system, and activation of the coagulation system. These immune responses can serve as triggers for severe infections, including COVID-19, and other infectious diseases. In our current project, known as the DIYUFOOD project, we aim to leverage extensive multi-omics and clinical data collected from four population-based cohort studies conducted in Israel, the Netherlands, the UK, and China. These studies encompass diverse ethnic backgrounds, dietary patterns, and socioeconomic statuses. By employing mediation analysis in observational studies and Mendelian randomization (MR) analysis, which uses genetic variants as instrumental variables for causal inference, we seek to unravel the causal pathways involved.

Initially, we will explore the potential associations between dietary patterns and immune response, as well as identify potential metabolic mediators that can explain the observed connections between diet and immune response. This analysis will be conducted in the four observational studies. Next, we will investigate the mediating roles of gut microbial composition in the associations between dietary patterns and immune response. Once specific species of microorganisms are identified in this second step, we will further examine the roles of circulating metabolites in mediating the associations between gut microbial composition and immune response. It is important to note that the gut microbiota produces an incredibly diverse range of metabolites. Finally, for all the associations observed from step 1 to step 3, including those between metabolic syndromes, gut-microbiota derived metabolites, and immune response, we will utilize MR studies to further validate the causal relationships. Moreover, we will expand our analysis to explore the associations between metabolic syndromes, gut microbiota-derived metabolites, and the severity of COVID-19 as well as long-COVID. This will be accomplished through the utilization of multivariable MR analysis, which takes into account multiple variables simultaneously.

The biological "food path" from dietary intake to immune response and subsequent disease outcomes is highly complex. Our project aims to provide a deeper understanding of the underlying causal pathways, as this knowledge is essential for developing effective food solutions to mitigate disease risks.

The DIYUFOOD project, initiated in May 2023, is currently focused on harmonizing dietary patterns to ensure consistency and comparability across various studies. To facilitate statistical analysis of dietary patterns, our primary objective is to standardize food groups by creating a unified framework for categorizing foods across different questionnaires. This involves mapping specific food items to broader food groups or categories that are consistent across all questionnaires involved in the project.

One of the key strengths of our project lies in the diverse composition of the participating cohorts, encompassing individuals with varying ethnic backgrounds, dietary patterns, and socioeconomic statuses. Leveraging this advantage, our initial project aims to systematically evaluate the validity and reproducibility of statistical methods for dietary pattern analysis across these diverse cohorts. This evaluation is particularly important as it addresses a significant gap in the existing literature, where such validation studies are limited.

The evaluation of validity will be conducted from two distinct aspects. Firstly, we will assess the validity of identified dietary patterns by comparing them to the true underlying dietary patterns within each cohort. This analysis will provide insights into how accurately the statistical methods capture and represent the actual dietary patterns observed in the populations studied. Secondly, we will evaluate the validity of the strength of associations between the identified dietary patterns and health-related outcomes. This assessment will enable us to determine the reliability and consistency of the observed associations between dietary patterns and various health outcomes.

The findings from these validation studies will significantly contribute to the understanding of dietary pattern methods and strengthen the overall inference derived from studies utilizing dietary patterns. By enhancing the scientific rigor of dietary pattern analysis, our project aims to improve the reliability and validity of future research investigating the associations between dietary patterns, health outcomes, and disease risks.