Danish researchers have used supercomputers to create an extensive map of protein complexes involved in diseases such as breast cancer, sclerosis and Alzheimers. These results aim to ease the discovery of proteins causing disease, and hence improve treatment.
The results show that proteins important for the development of different dis-eases have a tendency to form similar complexes.
This discovery was made by letting supercomputers at the Danish Center for Scientific Computing (DCSC) analyse hundreds of the many thousands known interactions, as well as texts written by disease experts, describing the different disease characteristics. Hereby it was possible to evaluate the similarity of differ-ent diseases, and thus create a so called human "phenome network" between the diseases. This work required vast amounts of computer resources, but has now led to a first draft of more than 500 protein complexes, associated with human pathology.
Furthermore, the research group trained a computer, a "learning-based predictor", to prioritize genes associated with disease. This lead to the identification of 114 new genes, which are very likely to be involved in various human diseases. These genes are now targets for drug development.
The Computational Science aspect of the work is especially relevant in a Danish setting, where hundreds of high-quality Bio-banks with patient material have been collected during the last 50 years. This type of eScience based bio-research can well turn out to be of significant value to global health as well as of major economic value to the Danish export marked.
Read more about the research: