Protein preservation has been a long lasting research topic due to its importance in many bio-pharmaceutical applications. A ”cold chain” is a commonplace solution to protein preservation, which stores biochemical products at a refrigerated temperature. A big advantage of cold chain is that the storing process is straightforward, without many further processes before the use of stored bio-products. However, it can also experience malfunction of the cooling system and results in economic lost and health care crisis. Ionic liquids (ILs), as a type of non traditional solvents, consist only of ions and are reported to be a potential candidate to replace the use of cold chain. The advantages of ILs include low flammability, high conductivity and less toxicity compared to some organic solvents. The most interesting feature of ILs is their extremely large number of cation-anion combinations, that can be tailored for specific use according to different needs. This thesis aims to investigate specific mechanism behind how ILs modulate protein behavior, specifically, how ILs affect protein stability, activity, and aggregation. We approach the research questions through the lens of molecular dynamics (MD) simulations and complement with experimental findings. In the first part of the thesis we first investigate the effects of two imidazolium based ILs (1-ethyl-3-methylimidazolium ethylsulfate, [EMIM]+[EtSO4]− and 1-ethyl-3-methylimidazolium diethylphosphate, [EMIM]+[Et2PO4]−) on lysozyme stability and activity. We collaborate with an experiment group at the University of Massachusetts (Bermudez lab) to complement our simulation results. Both ILs are found to destabilize lysozyme stability. In addition, both the cation and anions lower the stability of lysozyme, but in a different fashion. [EMIM]+ interacts with an Arg-Trp-Arg bridge that is critical in lysozyme stability through π–π and cation–π interactions, leading to a local induced destabilization. On the other hand, both anions interact with the whole protein surface through short-range electrostatic interactions, with [Et2PO4]− having a stronger effect than [EtSO4]−. Lysozyme activity is also reduced by the presence of the two ILs, but can be recovered after rehydration. It is found that the protein-ligand complex is less stable in the presence of ILs. In addition, a dense cloud of [EMIM]+ is found in the vicinity of the lysozyme active site residues, possibly leading to a competition with the sugar ligand. A fast leaving of these [EMIM]+ is observed after rehydration, which explains the reappearance of the active site and the recover of lysozyme activity. Although classical all-atom MD simulations can provide us with a great deal of microscopic information, they are often limited by the temporal-spatial scale of the simulated systems. For example, systems with high viscosity solvents or systems involving large number of atoms will be difficult to reach convergence for all-atom MD. In this case, coarse grained (CG) MD can come into play to achieve the desired time- and length- scales. The faster sampling obtained from CG MD is achieved by reducing the degree of freedom of the system and by removing local energetic barriers. In CG MD, similar atoms are grouped to functional groups and thus the free energy landscape is smoothen. We develop a novel CG MD named ”Protein Model with Polarizability and Transferability (ProMPT)”. The novelty of this model is the inclusion of the charged dummies that can result in change of dipoles. These dipoles can reflect the change of environments and thus allow the model to respond to different environmental stimulus. We validate ProMPT with several benchmark proteins: Trp-cage, Trpzip4, villin, ww-domain, and β-α-β. ProMPT is able to simulate folding-unfolding and secondary structure transformation with minimal constraints, which is not feasible with previous CG models. In addition, ProMPT can also reproduce the experimental results for the dimerization of glycophorin A (GpA) with different point mutations. Here we demonstrate the ability of the model to capture the change of conformational space caused by point mutation. In the last part of this thesis, we combine ProMPT and an in-house CG IL model to study the effects of [TEA]+[Ms]− on amyloid beta 16-22 (Aβ16−22) aggregation. Aβ16−22 is the hydrophobic core region and is the smallest fragment of Aβ that can fibrilize. Aβ has been extensively linked to the pathogenesis of the Alzheimer’s disease. [TEA]+[Ms]− is reported to suppress the formation of β-sheets and induce helices at high concentration. From our results, both β-sheet content and the aggregate size decrease with the increase of IL concentration, which are in agreement with experiments. Aggregates can form in both water and IL, but with different morphologies. In water, a nice hydrophobic core involving Phe-Phe interactions can form as well as intact β-sheet contacts. In addition, a cross β-sandwich structure is also observed, as seen from previous literature. However, the same hydrophobic core can not persist in the presence of IL. Aggregate structures in IL are not stable over time due to the [TEA]+-Phe interaction. Helicity is also computed for Aβ16−22 in water and in IL at different concentrations and a positive correlation is found. The increase in helicity at high [TEA]+[Ms]− concentration can be explained by the reduction of the inter-peptide contacts, which then increases the opportunity for the peptides to form helical structures. Single peptide studies also reveal that [TEA]+[Ms]− increases the helicity, possibly through cation-induced dipole enhancement. In this thesis, a series of detailed investigations on the effects of ILs on protein behavior is performed. Specific interactions between IL functional groups and protein local/global structures are examined. The mechanisms we studied here will help constructing a holistic view for the design of IL-protein pair applications. The construction of the new CG protein/IL model provides another tool for the scientific community to study secondary structure transformation, folding- unfolding, and other biochemical processes that are sensitive to the environment with CG MD.