Lay Description
This project aims to deepen our scientific understanding of diseases and their mechanisms. By studying genetic information alongside detailed clinical data e.g. symptoms and outcomes, researchers hope to discover new insights into the causes of diseases (“mechanistic insight”). A primary goal is to improve “diagnostic uplift,” which means finding accurate molecular causal diagnoses for patients where their symptom are likely to have an underlying genetics cause – but this is not currently known. This can be the case for patient with rare diseases and inherited or sporadic cancers. Understanding the molecular genetic cause of disease or cancer helps in the development and personalised prescribing of medicines tailored to target the cause of ill health at an individual level. Additionally, the project intends to explore “pharmacogenomics,” examining how genetic variations influence how patients respond to medications, including hypo or hyper response and/or potential adverse reactions.
The study will focus on patients from University Hospital Southampton (UHS) who have been referred for genomic testing within the NHS. This referral will have been made by the individual’s clinical team because of high suspicion of a genetic cause or driver. Researchers will securely extract genomic data (DNA information) and link it with “longitudinal clinical outcome data” (detailed medical records tracked over time) within the Wessex SDE.
The team will use statistical and mathematical modelling tools, test and validate approaches using artificial intelligence (AI) including decision support software, to analyse this combined data. They plan to use standardised medical terms (e.g. ICD10 codes, SNOMED CT, Human Phenotype Ontology (HPO)) to describe patients’ symptoms in a standardised format aligned with FAIR data principles. Using the SDE will enable granular clinical data to be sourced from across the range of clinical specialities and better connect a patient’s genetic makeup (“genotype”) with their clinical manifestations and symptoms (“phenotype”). This can be particularly important for patients with syndromes requiring care from across many clinical specialities e.g. a child with neurodevelopmental delay who has congenital heart defects and deafness may have important clinical details recorded within neurology, cardiology and audiology. Gathering the list of symptoms from across specialities in a standardised form and integrating with genomic data, is likely to increase the chance a causal diagnosis can be found. Current standard of care means that patients are often referred for genetic testing by one speciality and details provided in the test referral are often limited to that time point and that speciality. This limits the interpretive value of expensive genetic data that is generated for that patient and results in a negative report.