Overcoming Complexity of Biological Systems: from Data Analysis to Mathematical Modeling

摘要

The problem of dealing with complexity arises when we fail to achieve a desired behavior of biological systems (for example, in cancer treatment). In this review I formulate the problem of tackling biological complexity at the level of large-dimensional datasets and complex mathematical models of reaction networks. I show that in many cases the complexity can be reduced by using approximation by simpler objects (for example, using principal graphs for data dimension reduction, and using dominant systems for reducing complex models). Examples of dealing with complexity from various fields of molecular systems biology are used, in particular, from the analysis of cancer transcriptomes, mathematical modeling of protein synthesis and of cell fate decisions between death and life.