Enhanced symbolic regression to infer biochemical network models
July 14-18, 2019
Biological systems can be represented as complex networks illustrating the relationships and connections among biochemical species. Complex networks can uncover vital information regarding specific pathways or network bottlenecks, helping to reveal novel discoveries relevant to a variety of applications. Biological networks, however, are often highly interconnected and non-linear in nature making development of a comprehensive model challenging. Large amounts of data can be acquired to elucidate specific pathways, but deducing the entire network topology requires more rigorous computational techniques. There are in silico techniques, including evolutionary algorithms, to predict network topologies using information from experimental data. Biological networks can be decomposed into a system of differential equations under mass action kinetics assumptions describing the rate of change of the various biochemical species in the network. Symbolic regression can be used to generate a system of equations from acquired data.
Please click Additional Files below to see the full abstract.
Nicole Beauregard, Mark Marten, Steven Harris, and Ranjan Srivastava, "Enhanced symbolic regression to infer biochemical network models" in "Biochemical and Molecular Engineering XXI", Christina Chan, Michigan State University, USA Mattheos Koffas, RPI, USA Steffen Schaffer, Evonik Industries, Germany Rashmi Kshirsagar, Biogen, USA Eds, ECI Symposium Series, (2019). https://dc.engconfintl.org/biochem_xxi/26