Michael Blinov Post-Doctoral Research Associate Theoretical Biology and Biophysics Group (T-10), MS K710 Los Alamos National Laboratory, Los Alamos, NM 87545 (505) 667-4111, office | (505) 665-3493, FAX | Phonebook | Google | Library Email: mblinov@lanl.gov Publications    |    CV   (pdf-file)   |    Software Courses taken at the Weizmann Institute

Ongoing research

• Mechanistic modeling of epidermal growth factor receptor (EGFR) signaling. The purpose of the work is to create the most detailed mechanistic model of EGF signaling. As a first step, I've created an extended model based on interactions described by Kholodenko et al. (1999) and used by Schoeberl et al. (2002) and several other research groups. The original model tracks the dynamics of 18 species that arise from interactions among five proteins: EGFR, the ligand EGF, the adapter proteins Grb2 and Shc, and the guanine nucleotide exchange factor Sos. However, many molecular species that are monitored experimentally are omitted from this model, e.g. different phosphoforms of EGFR, or complexes involving several adapter proteins bound to receptor-dimer.Indeed, the model is based on several implicit assumptions: 1) the only monomers of EGFR considered are those lacking cytoplasmic modifications; 2) dimers of EGFR may not be in direct contact with more than a single adapter protein; 3) phosphorylation and dephosphorylation is simultaneous on both receptors in a dimer and all domains within a receptor; 4) a single protein bound to a reeceptor protects all the sites in a dimer from dephosphorylation; and 5) phosphorylated dimers cannot break up. There are no data justifying these assumptions, and some data contradicts them. For example, Jiang and Sorkin (2002) have shown that the EGFR dimer may be in contact with more than one adapter protein. When we lift these assumptions, we obtain a model with 356 distinct chemical species and 3,749 reactions. This model generates a variety of new predictions, e.g. time courses of phosphorylation for different tyrosines, and suggests experiments that can be performed to test these predictions.
• Development of general-purpose software BioNetGen for rule-based modeling of signal transduction based on the interactions of molecular domains. I have designed and implemented the first version of BioNetGen that was publicly released (reported in first-authored publications in Bioinformatics, 2004 and Complexity, 2005; a patent application is filed by the Los Alamos Natl Lab) and used to model different signaling systems. In this approach, rules for protein-protein interactions are specified and then used as generators of species and reactions. I used this software to study initial steps in signaling by immune-receptors (coauthoring publications in J. Immunology, Biotechn. Bioeng and Syst. Biol.), and developed test models for Toll-like receptor 4, interleukin 1, insulin, interferon gamma, and erythropoietin (Epo) receptor signaling.
• Currently, I'm developing advanced methods and software for rule-based modeling using graphs to specify biomolecular interactions (reported in LNCS, 2005 ), in order to explicitly and systematically represent the connectivity of components in multicomponent biomolecular complexes; and explicitly track and account for the internal structures of biomolecules. This progress allows me to model a large variety of different signaling systems, including systems where certain signaling complexes can form chains of different length.

Personal

	Operated by the University of California for the National Nuclear Security Administration, of the US Department of Energy. Copyright © 2001 UC | Disclaimer/Privacy