In the same way that the human genome project created invaluable genomic maps, the objective of this project is to develop methods for eventual construction of comprehensive genetic and metabolomic by microbome relationship maps. Such maps would be an invaluable resource for improving our understanding as to the underlying mechanisms by which microbes and ?omics features influence human diseases and conditions, potentially leading to identification of novel therapeutic targets. To these ends, this proposal seeks to develop statistical and computational tools for mapping associations and interactions between microbes and other ? omic features and for further utilizing other ?omics to improve microbiome based prediction models. Specifically, motivated by studies examining the role of the vaginal microbiome and other ?omics in birth outcomes and menopause, we aim to develop statistical methodology for (1) mapping genetic variants that influence microbiome composition so as to understand the innate component of the microbiome as well as learn mechanisms by which genetics influence outcomes; (2) creating global metabolic maps integrating both microbes and metabolites which will enable understanding of how perturbations might influence the system and identify key pathways for therapeutic target; (3) exploiting other ?omics in constructing more accurate microbiome based prediction models for preterm birth; (4) developing, distributing and supporting software packages for the proposed methods. The methods are based on frameworks in which we have considerable experience, but novel technical contributions are made to accommodate features of the data such as population stratification and relatedness in genetics, phylogenetic structure, and compositionality, as well as practical considerations such as availability of samples and other ?omics data. Consequently, these new methods have the potential for accelerating mechanistic and translational microbiome studies, developing vital resources for enabling systematic achievement of many biological, clinical, and public health problems that have eluded researchers for decades.
The methods developed in this proposal will enable improved understanding of the interactions between microbes and other ?omics, thus aiding in elucidation of the mechanisms by which microbes and ?omic features influence health outcomes and aiding in identification of potential molecular targets. Further emphasis is placed on utilization of other ?omics to develop microbiome based prediction models in pregnancy outcomes, improving early detection of women who are at risk of preterm delivery.
