<div>Computational Chemistry has proven to be an effective means of solving chemical problems. The two main tools of Computational Chemistry - quantum mechanics and molecular mechanics, have provided viable avenues to probe such chemical problems at an electronic or molecular level, with varying levels of accuracy and speed. In this work, attempts have been made to combine the speed of molecular mechanics and the accuracy of quantum mechanics to work across multiples scales of time and length, effectively resulting in simulations of large chemical systems without compromising the accuracy.</div><div><br></div><div>The primary tool utilized for methods development and application in this work is the Effective Fragment Potential (EFP) method. The EFP method is a computational technique for studying non-covalent interactions in complex systems. EFP is an accurate \textit{ab initio} force field, with accuracy comparable to many Density Functional Theory (DFT) methods, at significantly lower computational cost. EFP decomposes intermolecular interactions into contributions from four terms: electrostatics, polarization, exchange-repulsion and dispersion.</div><div><br></div><div>In the first chapter, the possibility of applying EFP method to study large radical-water clusters is probed. An approximate theoretical model in which the transition dipole moments of excitations are computed using the information from the ground state orbitals is implemented.</div><div><br></div><div>A major challenge to broaden the scope of EFP is to overcome its limitation in describing only small and rigid molecules such as water, acetone, etc. In the second chapter, the extension of EFP method to large covalently bound biomolecules and polymers such as proteins, lipids etc., is described. Using this new method, referred to as BioEFP/mEFP, it is shown that the effect of polarization is non-negligible and must be accounted for when modeling photochemical and electron-transfer processes in photoactive proteins.</div><div><br></div><div>Another area of interest is the development of novel drug-target binding models, in which a chemically active part of the ligand is modified via functional group modification, while the rest of the system remains intact. In the third chapter, the development and application of a drug-target binding model is explained.<br></div><div><br></div><div><div>Lastly, in the fourth and final chapter, we show the derivation for working equations corresponding to the coupling gradient term describing the dispersion interactions between quantum mechanical and effective fragment potential regions.</div><div><br></div><div>The primary focus of this work is to explore and expand the boundaries of multiscale QM/MM simulations applied to chemical and biomolecular systems. We believe that the work described here leads to exciting pathways in the future in terms of modeling novel systems and processes such as heterogeneous catalysis, QSAR, crystal structure prediction, etc.</div></div>