Sample scripts for the framework described in "Building up multiomics networks to decipher the role of gut microbiota in nutritional intervention trials"
Abstract: The availability of omics data providing information from different layers of complex biological processes, that link nutrition to human health, needs the development of integrated approaches combining holistically individual omics data, including those associated to the microbiota that impacts on the metabolization and bioavailability of food components. Microbiota must be considered as a set of populations of interconnected consortia, with compensatory capacities to adapt to the different nutritional intake. The consortium nature of the microbiome must rely on specially designed data analysis tools. The purpose of this work is to propose the construction of a general correlation network-based explorative tool, suitable for nutritional clinical trials, integrating omics data from faecal microbic taxa, stool metabolome (1H NMR spectra) and GC-MS for stool volatilome. The presented approach exploits a network-based descriptive paradigm necessary for a true multiomics integration of data, which is a powerful tool to investigate the complex physiological effects of nutritional interventions.
Code: Scripts for 1H NMR spectral feature processing, datasets merging (script: process_NMR_and_omicmerge), network construction and k-clique communities analysis (script: omic_networks_CNA)
Visualization of the framework, from data collection to network analysis.
Requirements: code is compatible with Python 3.8 environments, requiring the following packages to be properly installed: Numpy, Pandas, SKlearn, NetworkX