DTU Bioinformatics is structured in three scientific research themes revolving around the life science supercomputing infrastructure Computerome. The three research themes cover areas where we expect significant future developments as a response to large societal and industrial challenges. Human and animal health and diseases are part of the Health Informatics theme, whereas Biotechnology and Meta-genomic Informatics takes on the challenges and opportunities from the biotech industry as well as the genomic variation in all three domains of life. The third theme is Artificial Intelligence in Life Science, where we will build on the strong and innovative history of Center for Biological Sequence Analysis (CBS) within machine learning and expand into life science big and secure data.
Machine learning is a specialized field within artificial intelligence covering algorithms with the ability to learn (from data) without being explicitly programmed. In contrast to conventional top down hypothesis driven research, with machine learning techniques, biological mechanisms are identified directly from data most often without any prior hypotheses. Within biology this approach has proven very powerful. A focus of the department is thus to design and develop new algorithms and tools to meet the challenges of the amount and the complexity of big data within the health sector, and the upcoming data avalanche from the biotech industry.
Within the theme Health Informatics DTU Bioinformatics is engaged both in a) human genetics based health informatics, examples being the hologenetic view on childhood leukaemia and patient stratifications with traditional medicine approaches b) infectious diseases informatics, examples being the preparing for the next generation of global disease surveillance and building local outbreak predictors.
Biotechnology and MetaGenomic Informatics
Within the theme Biotechnology and MetaGenomic Informatics DTU Bioinformatics has strong interaction with the Danish Biotech industry in the discovery and refining of industrially important processes, examples being using large scale omics together with machine learning concepts to improve food related processes like wine fermentation and milk production.