Data based system design and network analysis tools for chemical and biological processes

Abstract

Analysis of Chemical and Bio Systems (ChemBioSys) of recent interest posses challenges such as unknown design principles, presence of non-linear dynamics, large scale of variables describing the system, different types and sizes of data, all create hurdles for systems analysis. The objective of this research is to understand the limitations of existing analysis methods and develop alternate tools/techniques necessary to solve problems related to such complex systems. Major research issues addressed include, dimensional reduction for optimum regression/classification, classification of non-linearly separated patterns, network synthesis and complexity analysis of networks. Important chemical process related problems like chemometrics, batch process monitoring and fault detection are attempted along with issues related to biological systems like cancer tumor classification, protein structure prediction, protein-protein interaction prediction, biological networks analysis. New system design and analysis concepts are proposed and implemented to resolve important ChemBioSys problems. The techniques are benchmarked with other existing techniques and advantages in terms of performance, generalizability and computational efficiency are established.