An integrative approach for understanding the adverse outcome pathways in algae

This project will use innovative Systems Biology approaches - including multiple 'omics' approaches (transcriptomics, metabolomics and lipidomics) and statistical modelling - to extensively characterise and then model the molecular and phenotypic responses of algae to stressors. Specifically, the project will develop models to predict the degree of activation of Adverse Outcome Pathways in algae in response to chemicals as representative stressors. The key objectives of the project are to:
1. Explore an available knowledge base to select a panel of relevant chemicals with both baseline and specific Mode(s) of Action (MoA).
2. Develop a dataset representing key (apical) endpoints for all chemicals selected in task 1.
3. Develop time-course gene expression, metabolomics and lipidomics profiling datasets representing algal cell molecular responses to chemical exposure as a representative stressor.
4. Construct gene regulatory networks and metabolic networks, encapsulating both the temporal changes and dose-response surface, to help to identify the key events in the stress pathway, adaptation to the stressor and potentially the "tipping points" to adverse (apical) endpoints within an AOP framework. Specifically, based on established biological network reconstruction methods [Becker & Palsson, Context-specific metabolic networks are consistent with experiments. PLoS Comput. Biol. (2008); Margolin, ARACNE: An Algorithm for the Reconstruction of Gene Regulatory Networks in a Mammalian Cellular Context. BMC Bioinformatics (2006)], we will use high quality genome-wide metabolic network reconstructions such as by Chang [Metabolic network reconstruction of Chlamydomonas offers insight into light-driven algal metabolism. Molecular Systems Biology (2011)] together with our omics datasets to construct temporal and condition specific metabolic and gene regulatory networks. We will apply biological network analysis algorithms, including our DiME algorithm (S. He, H. Chen, Z Zhu, D.G. Ward, H.J. Cooper, M.R. Viant, J.K. Heath, X. Yao, Robust twin boosting for feature selection from high-dimensional omics data with label noise. Information Sciences 291, 1-18 (2015) to analyse these temporal networks in order to identify the dynamics of pathways or network modules (cross-talk of pathways) perturbed by chemicals which exhibit adverse effects.