The Data Integration pillar will harmonize and integrate the multi-omics host-microbiome-nutrition data generated in the three biological pillars. This pillar will provide PIG-PARADIGM a unique strength to holistically model the host-microbiome holobiont in the developing pig. While leveraging the existing scientific domain-specific bioinformatics expertise in the biological pillars and the corresponding research groups generating the datasets, this pillar will also complement it with deeper (bio)statistical modeling expertise, which will augment the state-of-the-art analysis with novel statistical and machine learning approaches for data integration. It will develop novel graph-theory-based algorithms such as n-partite graphs and deep learning algorithms such as autoencoders to integrate host-microbiome multi-omics data and derive mechanistic insights.
The Data Integration pillar brings together (1) bioinformatics experts in the three scientific pillars, (2) experts in host-microbiome data integration and (3) experts in statistical modeling and data science. This first-of-a-kind synergy between bioinformaticians, data scientists and biologists to analyze and interpret big data will lead to novel molecular insights that can improve intestinal health of the developing pig.
The Data Integration pillar are organized into four data integration research themes (DRTs):