They develop a platform that will use AI to identify variations in the genome associated with diseases

VALENCIA, July 26 (EUROPA PRESS) -

A team of researchers from the Valencian University Institute for Research in Artificial Intelligence (VRAIN) of the Universitat Politècnica de València is carrying out the prototype of the Delfos platform that, through Explainable Artificial Intelligence (IAE), seeks to identify variations in the genomes that cause diseases, storing them in a data source that integrates all the information currently available in such a dynamic and constantly evolving environment.

The platform efficiently combines existing techniques and information dispersed in different sources. Its objective is to inform professionals and patients of those diseases that they can develop depending on their particular genome, why they are going to develop them and what are the applicable solutions with current medicine, always in accordance with the most personalized information that exists, the genome. individual.

The database has been designed to be continuously updated, incorporating discoveries associated with new genomic variations and their clinical significance, explains the academic institution in a statement.

This platform, developed by the PROS de VRAIN research group of the UPV, together with experts from the La Fe Health Research Institute (IIS La Fe), provides an Information System for the management of genomic variants, with the purpose of carrying out a proof of concept (PoC) of the Delfos platform that makes its subsequent translation into daily practice viable.

Currently, Delfos is practically at a level of technological maturity TRL7, since different laboratory experiments have been carried out at an academic level that mean that its development and validation phase can now be undertaken with the support of clinical experts from Pediatric cancer and family heart disease teams from IIS La Fe.

"Integrating this information, together with the set of clinical data of each patient, is undoubtedly the next step to establish precision medicine as a reality in clinical practice. In this way, we will be able to both anticipate the development of diseases and identify the most effective therapies by knowing in detail the genomic origin and the possibilities of response to drugs", points out the main researcher of this project at VRAIN-UPV, Óscar Pastor, who adds how "all of this will save costs, both in diagnosis and in treatments, and it will undoubtedly have an impact on the quality of life of patients.

The genomic analysis of a person can be preserved and reused, so that if, as Óscar Pastor explains, new variations continue to be discovered and "if a new relationship is discovered between said variations and a disease, and the genome of that patient is located in Delphi, it can inform you of the new knowledge generated when variations that in principle did not have clinical significance, at a certain moment have it.Therefore, it can be adapted to the information on the genome that is in continuous evolution and in which new discoveries follow one another", he highlights.

The platform has been developed with Explainable AI, which through a holistic conceptual model of the genome in which all the relevant parts of genomic knowledge are connected, complements predictive knowledge with an explanatory one. It not only predicts whether the disease is present, but also sets out to explain why it is.

The Delfos project, financed by the State Research Agency (subsidized by Next Generation EU, Ministry of Science and Innovation), will conclude in December 2023 and will allow its transfer to an industrial environment to begin.