The project focused on a highly targeted approach towards multi-drug therapies in cancers, with particular research on breast, ovarian, and prostate cancer. Our approach aimed at controlling cancer-specific essential genes, currently made available by the new CRISPR gene editing technology, as acting upon these essential genes is guaranteed to kill the diseased (and only the diseased!) cells. Our new machine learning algorithms identified nodes targetable by approved drugs, which lead to controlling essential genes, through (sometimes many) cascading effects in the network.
The project involved a consortium of 4 industrial partners, Orion Pharma, Misvik Biology, MediSapiens, and Euformatics, and one academic center, Åbo Akademi, i.e. our research group, as project coordinator.
- We generated a software tool, NetControl4BioMed, which is a pipeline for biomedical data acquisition and analysis of network controllability
- We proved that the structural target controlability problem, an important problem in the network control theory, is NP-Hard, namely is it a hard computational problem which needs to be approached by use of approximation and heuristic algorithms.
- We provided several approximation algorithms for the structural target controlability problem.
- We proved that the structural target controlability problem is fixed parameter tractable when parameterized by the number of target nodes