Aarhus University Seal / Aarhus Universitets segl

Rereading the genome

  • Title: Rereading the genome: applying natural language processing to improve genetic prediction of diseases
  • AU project manager: Professor Doug Speed, Center for Quantitative Genetics and Genomics
  • Collaboration partners: Søren Østergaard, Department of Clinical Medicine, Aarhus University
  • Project period: November 2020 - October 2022
  • Funding: 1,850,000 DKK

Project summary:
Many psychiatric and neurological diseases are highly heritable (e.g., schizophrenia, major depression, Ischemic Stroke and Alzheimer’s Disease all have heritability between 40 and 80%), and therefore, it should be possible to accurately predict which individuals will develop them based on genetic information. However, at present this is not the case. For most complex diseases, the best prediction models have accuracy less than a fifth the theoretical maximum. In a attempt to move the field of personalized medicine forward, we will adapt state-of-the-art tools for natural language processing (NLP), a branch of artificial intelligence, for use with genetic data.