Project 42
Project 42
In spite of a huge investment in both human and financial capital, the incidence of obesity and metabolic diseases, such as type 2 diabetes, is increasing worldwide. Although such diseases are rightly considered to be “complex”, the implications for how they should be studied and analyzed have not been fully appreciated. Their complexity is manifest in the enormous multi-factoriality associated with their aetiology, with factors ranging from the “micro”, such as genetic and epigenetic factors, to the “macro”, where public policy and socio-economic factors play a key role. The fact that these factors span a large spectrum of scales means that they span a large number of disciplines, such as genetics, physiology, psychology, sociology, economics and politics, to name just a few. From a data perspective, the multifactoriality of such diseases means that they have an enormously rich phenomenology that requires data from multiple perspectives be collected, integrated and analyzed to obtain a more holistic view. In short, it is a question of having “deep” data as opposed to just “big” data, where by “deep” we mean that the data covers multiple scales and disciplines.
The goal of Project 42 is to develop the deepest data set possible for the study of obesity and metabolic disease and to make that data available to researchers worldwide, either directly or through a machine learning-based modelling platform. Currently, Project 42 has data from over 7,000 participants covering many thousands of variables, from the genetic to the social and psychological. The Conductome project provides the conceptual and theoretical framework with which this data is being analyzed. A key finding of the project is the identification and confirmation of hundreds of variables that are all predictive of obesity and metabolic disease and which must be integrated into a coherent, holistic view of the evolution of these diseases.