Predicting Protein-Protein Interactions from Protein Domains Using a Set Cover Approach
IEEE/ACM Transactions on Computational Biology and Bioinformatics (TCBB)
| Hi-index | 0.00 |
Biocomplexity is the study of the complex relationships among biological entities that are responsible for life. This dissertation addresses two important computational in biocomplexity: morphogenesis and protein-protein interaction networks. Morphogenesis is the development of multicellular organisms. I develop models, algorithms, and software that integrate the genetic regulatory network and physical or generic cellular mechanisms that explain how cells interact to form tissues. The genetic regulation is modeled by a combination of a rule-based state automaton and a set of partial differential equations (PDEs); the generic cellular mechanisms include cell adhesion, cell differentiation, cell growth, mitosis, secretion of morphogens, haptotaxis and chemotaxis. Protein networks are part of signaling, metabolic, and other biological pathways important to cells and organisms. We develop a new algorithm called Maximum Specificity Set Cover (MSSC) to predict protein-protein interactions. The predictions by MSSC preserve not only the topological characteristics of protein interaction networks, but also the protein co-expression. Our method outscores other prediction methods in quality.