Aetionomy – Organising Mechanistic Knowledge about Neurodegenerative Diseases for the Improvement of Drug Development and Therapy

IMI’s Call 8 is geared to develop knowledge frameworks for ‘druggable mechanisms’ for two domains of pathophysiology. In response to this call, we propose (I) the development of a generic AETIONOMY pipeline to capture and infer over mechanistic knowledge of pathophysiology, and  (II) the focused application of this pipeline to derive clinically significant mechanistic taxonomies of neurodegenerative diseases.

These taxonomies will represent multi-scale structural and functional understanding of the causal mechanisms of and possible treatments in PD and AD. In particular, knowledge integration about neural substrates and their functional correlates will be generated through a pipeline that co-ordinates:

• I. the manual knowledge acquisition of mechanistic neurological knowledge from clinical experts,
• II. the semi-automated calculation of functional indices derived from clinical examination datasets, neuroradiological images and clinical and experimental investigation results,
• III. the large-scale acquisition of further evidence about functional relationships between knowledge created in i & ii by mining electronic health records and the literature,
• IV. the automated detection of significant correlations inferred from knowledge derived from steps i to iii above.

To that end, AETIONOMY has assembled a seasoned consortium of expert clinicians and scientists to build upon (A) previous work within the consortium, which comprises unpublished and published work (ontologies such as ADO, the Alzheimer´s Disease Ontology, in press in “Alzheimer´s & Dementia”; disease models such as the comprehensive CellDesigner model for PD build by partner LCSB) and (B) considerable public domain resources generated by large international communities (e.g. the Neuroscience Information Framework), and (C) results from ongoing IMI projects (e.g. DDMoRe, OpenPHACTS, eTRIKS, EMIF), providing a unique combination of tools and expertise to discover the core criteria that make for a good drug target and, in so doing, classify patients according to these criteria.