The browser you are using is not supported by this website. All versions of Internet Explorer are no longer supported, either by us or Microsoft (read more here:

Please use a modern browser to fully experience our website, such as the newest versions of Edge, Chrome, Firefox or Safari etc.

Default user image.

Giuseppe Giordano

Research team manager

Default user image.

Discovery of drug–omics associations in type 2 diabetes with generative deep-learning models


  • R.L. Allesøe
  • M. Ridderstråle
  • G.N. Giordano
  • P.W. Franks
  • L. Groop
  • N.A. Pasdar
  • H. Fitipaldi
  • A. Kurbasic
  • P. Mutie
  • H. Pomares-Millan
  • M. Klintenberg
  • M. Abdalla

Summary, in English

The application of multiple omics technologies in biomedical cohorts has the potential to reveal patient-level disease characteristics and individualized response to treatment. However, the scale and heterogeneous nature of multi-modal data makes integration and inference a non-trivial task. We developed a deep-learning-based framework, multi-omics variational autoencoders (MOVE), to integrate such data and applied it to a cohort of 789 people with newly diagnosed type 2 diabetes with deep multi-omics phenotyping from the DIRECT consortium. Using in silico perturbations, we identified drug–omics associations across the multi-modal datasets for the 20 most prevalent drugs given to people with type 2 diabetes with substantially higher sensitivity than univariate statistical tests. From these, we among others, identified novel associations between metformin and the gut microbiota as well as opposite molecular responses for the two statins, simvastatin and atorvastatin. We used the associations to quantify drug–drug similarities, assess the degree of polypharmacy and conclude that drug effects are distributed across the multi-omics modalities. © 2023, The Author(s).


  • Genomics, Diabetes and Endocrinology
  • EXODIAB: Excellence of Diabetes Research in Sweden
  • Genetic and Molecular Epidemiology
  • EpiHealth: Epidemiology for Health
  • eSSENCE: The e-Science Collaboration

Publishing year





Nature Biotechnology

Document type

Journal article


Nature Publishing Group


  • Endocrinology and Diabetes


  • Deep learning
  • Learning systems
  • Patient treatment
  • 'omics'
  • Auto encoders
  • In-silico
  • Learning models
  • Multi-modal data
  • Multi-modal dataset
  • Non-trivial tasks
  • Omics technologies
  • Phenotyping
  • Type-2 diabetes
  • Modal analysis



Research group

  • Genomics, Diabetes and Endocrinology
  • Genetic and Molecular Epidemiology


  • ISSN: 1087-0156