Mathematical methods for ‘Reverse Engineering’
“Reverse engineering” Biological Networks. To understand nonlinear and highly interconnected signaling pathways in cancer cells, I am applying computational approaches that directly "reverse engineer" networks from large scale quantitative experimental data that I am collecting. Biological data are naturally noisy and sparse and therefore particularly suited to a probabilistic modeling framework, of which Bayesian network analysis is one of the best developed.When inputs and outputs are known, Bayesian modeling can be used to infer the functional interactions (Inference). When the inputs and the system are known, the model can be used to predict a response (Prediction). When the system and the desired output are known, the model can serve to design optimal modes of control (Control).
Drug Target Deconvolution. The ‘magic bullet’ or ‘one target, one drug’ notion of therapeutic development has proven futile. The emerging field of network pharmacology recognizes that to combat a disease effectively, a drug must target not one but several building blocks in a biological network. Designing drugs with a specific multi-target profile or a rational combination of such drugs is a complex and difficult task, but it has greater potential to improve the balance of efficacy and safety than work with single target agents. Thus, there is an increased incentive to develop new systems-based methods capable of exploiting the known polypharmacology of drugs in order to identify the molecular targets of active hits, also called ‘target deconvolution’.
In collaboration with Dr. Leon Peshkin, I am applying statistical machine learning-based approaches, combined with pharmacological response data to ‘deconvolve’ and identify kinases that are important for epithelial and mesenchymal cell migration. Our approach is based on the reality that while all kinase inhibitors have broad specificity, the breadth of each one is different. Screening a set of optimally designed kinase inhibitors in a panel of cancer cell lines, we identified cell-type specific kinases that regulate cell migration. Using this approach, we can also predict cell-type specific response to unseen kinase inhibitors.