In silico modeling of chemical and biological interactions at different scales

Identifier:  TDX:2321

Handle: https://hdl.handle.net/20.500.11797/TDX2321

Authors:  Kamath, Padmaja Balachandran

Abstract:
In the past decades, government, society and industry at large have taken keen interest in the impact at different scales that exposure to chemicals has on humans and environment. Many countries governments have imposed regulations as per which it has become important to establish the potential effects of these chemical entities with respect to human health and environmental endpoints. Given the time taken by traditional tests, costs and large number of chemicals to be evaluated, there has been a rapid growth in the number of computational models that link the structure of chemicals to their biological activity. To extend the basis of knowledge that currently exists in Structure Activity Relationship (SAR) models for chemicals, a similar approach was used to develop a new model and generate sets of metabolic triggers which can be used together with Q(SAR) methods. This thesis presents SAR rules for prediction of mutagenicity in vitro, along with metabolic triggers for prediction of mutagenicity in vitro and in vivo. Along with chemical compounds, nanoparticles are also being used increasingly across different classes of consumers’ products. Since, in physiological context, the protein corona constitutes the interface between the nanoparticle and cells, it plays a fundamental role in nanoparticle-cell association. In this thesis, the physicochemical properties of protein corona were used to develop a model to predict cell association. Lastly, this thesis focuses on the topic of drug resistance in bacteria, which has become a matter of global concern. With bacteria growing resistant to antibiotics at a faster pace than discovery of new antibiotics, information on the response that new bacterial proteins would have to the currently available antibiotics, based on their similarity with the known antibiotic-resistant proteins is necessary. An alignment-free method was developed to improve the resistance profile classification of bacterial proteins based on their physicochemical properties.