9 Exposomic studies

In this chapter, we study exposomic data. This type of data is different from the omic data that we discussed in the previous chapters. To start, the exposure variables that constitute the dataset are measured with different experimental techniques. They are highly diverse and heterogeneous, as they may include biological assays, obtained from personal questionnaires or derived from inferred exposures at geolocalized sites. Nevertheless, organisms respond to environmental changes and their efficiency in the response may result in phenotypic differences that may lead to adaptation or disease. The identification of the environmental changes is therefore essential to identify phenotypic differences among individuals and an unbiased scan on a large number of exposures is a starting point. In addition, exposure data may affect other molecular processes such as gene transcription or methylation. And therefore, important mechanisms may be discovered from the association between exposomic and transcriptomic, or methylomic data. Integration of omic data is then required. From the analysis perspective, exposomic data can be treated as the previous omic data, in which data structures are defined and methods are implemented on them. Here, we illustrate the use of rexposome, a comprehensive Bioconductor package with functions to impute, normalize and characterize correlational structures in exposomic data, to perform exposome-wise association analysis (ExWAS) and to test the association between exposomic and other omic variables.

The R code to reproduce the results and figures of this chapter can be found here.