Date of Award


Publication Type

Master Thesis

Degree Name



Computer Science

First Advisor

Luis Rueda


Pure sciences, Biological sciences, Applied sciences, Complex type prediction, Electrostatic energy, Machine learning, Pattern classification, Protein-protein interactions, Support vector, machines



Creative Commons License

Creative Commons Attribution-Noncommercial-No Derivative Works 4.0 License
This work is licensed under a Creative Commons Attribution-Noncommercial-No Derivative Works 4.0 License.


Prediction and analysis of types of protein-protein interactions (PPI) is an important problem in molecular biology because of its key role in many biological processes in living cells. In this thesis, I propose a model called PPIEE (Protein-protein interaction using electrostatic energies) to predict and analyze protein interaction types using electrostatic energies as properties to distinguish between these types of interactions. This prediction approach uses electrostatic energies for pairs of atoms and amino acids present in interfaces where the interaction occurs. Using this approach, the results on well-known datasets confirms that electrostatic energy is an important property to predict obligate and non-obligate protein interaction types. The classifiers used are support vector machines and linear dimensionality reduction. Since electrostatic interactions are long ranged, some other experiments are performed by changing the threshold values, which are the distances calculated between atom pairs of interacting chains, ranging from 7Å to 13Å. This information will be helpful for researchers to understand how different physiochemical properties contribute to understanding about stability of protein complexes and their function.