One major bottleneck in present biomedical research is to predict cell behavior and drug response. Until now, high-throughput data and biological knowledge have been mostly used to generate descriptive signaling diagrams of intricate molecular networks, which, however, are not sufficient to predict biological responses. The great challenge of next ten years is to transform these descriptive diagrams into functionally predictive dynamic models of molecular networks. To this end, we are tactically focusing on modeling the relatively small Ras/MAPK signaling pathway. In addition, as a long term objective, we aim to develop a multi scale computational model to simulate cell population growth and cell cycle progression.
The immediate objective is to construct a mathematical model of the SOS/Ras, MAPKs, and PI3K/Akt modules sufficient to mimic, through computational simulations, experimental results obtained with the mouse Y1 adrenocortical malignant cell line. In the mid term, we intend to predict the behavior of such modules in other cell lines, for instance, in human HEK293 tumor cells and immortalized human epidermal keratinocytes. Finally, the long-term objective is to construct a multi-scale, computational model of automata to simulate population growth and cell cycle progression of population of cells under stimulation of toxins, hormones, serum factors, and growth factors such as the Fibroblast Growth Factor 2 (FGF2).
To achieve the proposed objetives, this sub-project relies on the Center platforms, especially on the Group of Computational Biology and Bioinformatics (GC2B). Some of the current projects carried out by GC2B are of immediate interest of this sub-project, namely the development of the CeTICSdb platform for storage and analyses of heterogeneous omics data, and the SigNetSim framework for mathematical modeling and computational simulation of signaling network kinetics.
Initially, models are designed using low-throughput data from traditional molecular biology techniques (e.g., quantification of Western blot experiments). However, one mid-term objective of the Proteomics platform is to establish routine techniques for high throughput quantitative phospho-proteomics based on mass spectrometry (MS), which will allow us, in the near future, to monitor the kinetics of the SOS/Ras/MAPKs module along with parallel signaling pathways (e.g., PI3K/Akt) in different cell lines. Such high-throughput data will be stored in the CeTICSdb platform and employed to build predictive kinetic models with the assistance of the SigNetSim framework.
The Ras/MAPK module is a small network (~50 components) that plays central roles in cellular homeostasis. In addition, gene or oncogene amplification is a major genomic alteration underlying deregulation of signaling networks. Thus, the construction novel Ras/MAPK/PI3K/Akt models to mimic experimental results of different cell lines is a timely and feasible goal. On the other hand, the construction of cell cycle large-scale models is an ambitious and long term goal. Nonetheless, these two goals fit well with methodological demands of the Center research to convert descriptive molecular diagrams into predictive dynamical computational models.
Hugo Aguirre Armelin, Principal Investigator
Junior Barrera, Principal Investigator
Marcelo da Silva Reis, Associate Investigator
Milton Yutaka Nishiyama Junior, Associate Investigator
Matheus Henrique dos Santos Dias, Postdoctoral Fellow
Vincent Noël, Postdoctoral Fellow
Cecília Sella Fonseca, PhD Student
Eduardo Lopes da Silva, PhD Student