| 摘要 |
Presented herein are techniques and methodologies for creating large-scale data-driven models of biological systems and exemplary applications thereof including drug discovery and industrial applications. Exemplary embodiments include creating a core skeletal simulation (scaleable to any size) from known biological information, collecting quantitative and qualitative experimental data to constrain the simulation, creating a probable reactions database, integrating the core skeletal simulation, the database of probable reactions, and static and dynamical time course measurements to generate an ensemble of biological network structures and their corresponding molecular concentration profiles and phenotypic outcomes that approximate output of the original biological network used for prediction, and finally experimentally validating and iteratively refining the model. The invention further describes methods of taking conventional small-scale models and extending them to comprehensive large-scale models of biological systems. The methods presented further describe ways to create models of arbitrary size and complexity and various ways to incorporate data to account for missing biological information.
|
