The overall objective of the Ideker Laboratory is to develop an artificially intelligent model of the cell able to translate a patient's data into precision diagnosis and treatment. For this purpose, we run an experimental facility for mapping the gene and protein interaction networks that govern eukaryotic cell biology, with a focus on pathways underlying cancer and neurodevelopmental disorders.
- Bandyopadhyay et al. Science. 2010 Dec 3;330(6009):1385-9.
- Kramer et al. Molecular Cell. 2017 Feb 16;65(4):761-774.e5.
- Shen JP, Nature Methods. 2017 Jun;14(6):573-576.
Our major bioinformatics challenge is to model how cells process information from genotype to phenotype. Towards this goal, we develop machine learning methods that attempt to learn cell structure and function directly from genome-scale datasets:
- Ma J, et al. Nature Methods. 2018 Apr;15(4):290-298.
- Yu M, et al. Cell Systems. 2016 Feb 24;2(2):77-88.
- Dutkowski J, et al. Nature Biotechnology. 2013 Jan;31(1):38-45.
However, much remains to be done before we have a cell model capable of making robust predictions about patients. A recent breakthrough on this front was our creation of the D-Cell, a deep neural network modeling the inner workings of a eukaryotic cell. We are also developers of Cytoscape, a popular platform for visualization and modeling of biological networks which is supported by a consortium of many labs including our own (http://www.cytoscape.org/).