Precisionsdiagnostik för svåra lungsjukdomar baserad på DNA-metyleringssignaturer

Analyses of DNA methylation (DNAm) status forms the mechanistic basis for a number of diagnostic methods used for detection, treatment monitoring and decision-making in the cancer field.

Although a number of DNAm signatures have been described for non-cancer diseases, none of them has yet reached clinical use.

The aim of the project is to validate our DNAm signatures for severe infections including tuberculosis and COVID-19 identified through previous VR funding for use in lung medicine, to fill the portfolio with additional lung disease-related DNAm signatures and to create a classifier algorithm that can be used to identify lung diseases.

A multidisciplinary team consisting of pre-clinical researchers, binonformaticians, engineers, cilnicians and business experts has been created to address these tasks.

At the clinics for infectious disease and lung medicine at Linköping University Hospital, DNA will be isolated from lung patients´ sputum, mouth swab and fingerprick blood samples and analysed by applying artificial intelligence (AI)-based analysis tools to identify disease-relevant DNAm patterns linked to deidentified clinincal metadata and laboratory findings.

The AI approach will be developed towards a prototype classifier algorithm during the final months of the project.If successful, the project catalyzes the transition of DNA methylation biosignatures from bench to bedside. The versatility and adaptablility of such tools can contribute to future precision medicine.