Reform the taxonomy of human diseases.

In their 2011 Nature Reviews Drug Discovery paper “A call to reform the taxonomy of human disease” *, Ismail Kola and John Bell formulate an impressive plea for “A coordinated effort to incorporate advances in the understanding of the molecular and genomic variations in common diseases, such as hypertension, into their diagnosis and treatment could transform drug development and medicine.”

The two authors emphasize the need for a re-classification of diseases according to mechanisms that contribute to the aetiology of a disease at the molecular (“omics-“) level rather than the current phenotypic approach. They point out that several diseases that appear to be distinct at the clinical phenotype level may actually go back to one shared disease mechanism, or that diseases classified under one concept are in fact phenotypically similar manifestations of different, diverse disease mechanisms.

Kola and Bell underline that the concept of a mechanism-based taxonomy would also address the challenge of subgroup-specific or even individualised medicine, as an in-depth knowledge about the mechanisms underlying a disease inherently provides the basis for strategies to treat that disease in a dedicated, mechanism-specific way.

AETIONOMY is a response to the challenge they laid down and is a knoweldge generation approach which can accommodate the specific data types and data structures for a broad variety of indication areas whilst retaining the capacity to construct disease-specific taxonomies based on the underlying disease mechanisms. It is part of a larger, coordinated effort to establish a mechanism-based taxonomy which ultimately will result in improved diagnostic and therapeutic procedures.

*Ismail Kola & John Bell
A call to reform the taxonomy of human disease
Nature Reviews Drug Discovery 10, 641-642 (September 2011) | doi:10.1038/nrd3534

Concept

Data, disease modeling and reasoning.

The AETIONOMY concept foresees a primary role of the taxonomy in
i) describing and organising the indication-specific data in the data cube, in
ii) linking the data to disease models that are based on causal and correlative relationships and in
iii) support of reasoning over the knowledge that is explicitly represented in related ontologies or knowledge-based disease models.

The consortium has extensive and proven experience in the generation of disease-specific ontologies for NDDs, as demonstrated by the recent publication of the “Alzheimer´s Disease Ontology (ADO)”, and the generation, in collaboration with partners from the pharmaceutical industry, of disease ontologies representing substantial parts of the knowledge on Parkinson´s Disease, Multiple Sclerosis and Epilepsy.

AETIONOMY will not have the resources to validate the entire set of aetiologies linked to the taxonomy in the given time and within the budgetary limits. We have therefore carefully designed a validation strategy that will guide the final prospective clinical study meant to demonstrate the validity of the aetiology-based taxonomy. The consortium brings together four leading clinical centres with proven expertise in conducting such sort of studies; addressing effectively the need to validate the mechanism-based taxonomies for both, PD and AD. A dedicated AETIONOMY work package on ethical and legal aspects has a clear European perspective and scope and is set up in a way that reaches out beyond the AETIONOMY project and actively seeks the coordination with other projects funded under the same theme.

AETIONOMY makes extensive use of developments made in and funded by other IMI or EU projects. In the area of knowledge and data management, we build largely on the work done in OpenPHACTS; and we will re-use the entire data curation pipeline developed in the course of eTRIKS. Modelling and mining principles learned from VPH projects will guide our work, leveraging on our involvement in other large EU research initiatives. Finally, the substantial effort made on the side of clinical data integration in the course of EMIF, the European Medical Information Framework, will be accessible to AETIONOMY.