Metabolomics technologies, in particular mass spectrometry, are rapidly maturing and nearing the high throughput (HT) status, when one would be able to measure concentration of a large fraction of metabolites and other cellular components with high precision, high temporal resolution, and a relative low per- experiment cost. However, the experimental techniques, as well as data analysis protocols are still far from perfection. In particular, the sensitivity of HTP metabolic techniques still allows determination of concentrations of comparatively highly abundant species only. Thus the overall goal of this proposal is to take advantage of Los Alamos expertise in isotope labeling and analysis, as well as in reverse engineering (RE) of biochemical interaction networks from HTP data to develop a set of coherent experimental and computational tools and protocols for improving the quality of HTM measurements and analysis, and for using this data for reconstruction of small scale metabolic pathways, as well as system-wide metabolic networks. We will accomplish this goal through a series of three integrated Specific Aims. First, we will improve MS metabolic profiling by developing stable isotope-enhanced metabolome analysis methods, improving pathway tracing and improving peak detection software. Second, we will improve the unique reverse metabolic network inference software developed at Los Alamos by enhancing the structural data methods and improving HT analysis.
The third Aim i s an evolving series of experimental studies using in vitro models of tumor progression using standard 2-D and more complex 3-D cell culture systems. Not only will these experiments provide a rich dataset for testing and validation of the techniques developed in the first two Aims, but we have also designed our experiments to test a fundamental scientific hypothesis. Specifically, we propose that the profile of energy-related metabolites in mammalian cells is more affected by the stressful microenvironment found in solid tumors than by the initial transformation from a normal to a malignant phenotype.
These Aims are highly interactive, with experimental systems becoming more complex as the new MS tools advance in their development. In conclusion, we have designed a project that not only directly addresses the RFA and several key areas of the NIH Roadmap, but also will provide a large set of metabolic profiles to improve our understanding of the metabolomics of human cancer.
| Stromberg, Sean P; Antia, Rustom; Nemenman, Ilya (2013) Population-expression models of immune response. Phys Biol 10:035010 |
| Nemenman, Ilya (2012) Gain control in molecular information processing: lessons from neuroscience. Phys Biol 9:026003 |
| Bandaru, Pradeep; Bansal, Mukesh; Nemenman, Ilya (2011) Mass conservation and inference of metabolic networks from high-throughput mass spectrometry data. J Comput Biol 18:147-54 |
| Mu, Fangping; Unkefer, Clifford J; Unkefer, Pat J et al. (2011) Prediction of metabolic reactions based on atomic and molecular properties of small-molecule compounds. Bioinformatics 27:1537-45 |
| Cheong, Raymond; Rhee, Alex; Wang, Chiaochun Joanne et al. (2011) Information transduction capacity of noisy biochemical signaling networks. Science 334:354-8 |
| Otwinowski, Jakub; Tanase-Nicola, Sorin; Nemenman, Ilya (2011) Speeding up Evolutionary Search by Small Fitness Fluctuations. J Stat Phys 144:367-378 |
| Houston, Jessica P; Naivar, Mark A; Freyer, James P (2010) Digital analysis and sorting of fluorescence lifetime by flow cytometry. Cytometry A 77:861-72 |
