FindingPheno consists of eight work packages (WP) including:
- one management WP
- two structural WPs, and
- five technical WPs
Over 48 months the consortium will develop and implement novel technical and statistical methods, and provide access to easily accessible software packages using statistical methods, as well as undertake outreach and wide dissemination of project results.
UCPH lead WP5 to develop a hierarchical Bayesian framework that integrates existing biological knowledge with the omics data sets to create an inference model for predicting phenotype outcomes.
UCPH will also contribute novel structural causal models to WP3, adding directionality to the relationships identified by Machine Learning to give true causality (a causes b) rather than just associations (a and b are somehow related).
As the project coordinator, UCPH also leads Project Management (WP8) and Dissemination (WP1) with support from the consortium.
UTU leads WP4 where in close collaboration with CER they model the dynamic nature of the living system where the omic and meta-omic data changes over time or across different parts of the organism, and provides input into the other method development tasks in WP3 and WP5.
For this UTU brings their expertise in machine learning/AI, probabilistic programming, statistical ecology, and data science along with relevant experience modelling the dynamics of the human microbiome.
In WP4, CER will develop analytical and agent based models to study factors that determine host microbiome interactions. The aim is to develop models with increasing complexity to provide theoretical background to the statistical results received from the project's holo-omic data.
Based on data analysis, CER will also test new hypotheses on the evolution and ecological stability of host-microbiome communities by mathematical methods. This will provide a theoretical basis to the project and spearhead the generation of new hypotheses for the host genome-microbiome-phenotype axis in any biological system.
EMBL's leads WP2 where it serves as data coordinator. EMBL ensures uniform access to data and computing platforms across the project, and obtains genotype-phenotype associations on case study systems.
EMBL will implement an iterative feedback mechanism to provide validation and improvement of models developed in WPs 3 to 5 by applying the models to multi-omics data from the case studies.
Also, EMBL will develop a solution to enable the sharing of the project's data outputs and how these results may be consumed by relevant EMBL resources.
CF leads WP3 where it brings its expertise in the modelling and analysis of collective behavior as well as in the use of Machine Learning and AI methods to develop statistical models for multi- omics data.
Qiagen will lead WP7 and move FindingPheno-developed technology from TRL4 to TRL6.
Njorth Bio is the project's Innovation Manager, responsible for leading the commercial application of the project's innovations. This largely falls under WP1, where Njorth Bio with the Outreach Manager work to ensure the project's communication actions target the most relevant industrial end-users while also providing commercial training and advice to researchers involved in WPs 3 to 5.
In WP7 Njorth Bio-Sciences will test the project's new statistical tools against its in-house omics resources to validate the new solutions in a relevant environment.
Chr. Hansen will contribute to WP7, providing feedback for the development of a bioinformatic tool and as an end-user by focusing the project on business relevant questions across all food production sectors including crops, aquaculture and terrestrial livestock.