Skip to Main Content
 

Global Search Box

 
 
 

ETD Abstract Container

Abstract Header

Application of Molecular Simulations and Machine Learning Methods to Study Biological and Metallic Interfaces in Aqueous Environment.

Aghaaminiha, Mohammadreza
http://orcid.org/0000-0003-4748-3823
http://rave.ohiolink.edu/etdc/view?acc_num=ohiou162818080313617

Abstract Details

2021, Doctor of Philosophy (PhD), Ohio University, Chemical Engineering (Engineering and Technology).
Molecular dynamics (MD) simulation is a computational methodology to probe molecular-level details of physical systems. In MD, the motion of molecules is simulated and from the molecular trajectories, thermodynamic and kinetic properties are ascertained. Machine learning (ML) techniques are a set of computational tools that allow us to identify complex, non-linear relationships in data. ML methods are particularly useful when a system property of interest depends on a large number of variables, and there are no accurate physics-based models relating the property of interest to the variables. ML methods rely on the availability of large datasets to tease out the relationships between the variables. In this research, we have employed MD simulations to study the structural and thermodynamical properties of the simplified plasma membrane of eukaryotic cells, known as the lipid bilayer. We have applied ML methods to study the phase diagram of lipids. Furthermore, we have used ML methods for various corrosion-related applications. By studying asymmetric lipid bilayers using MD simulations, we have shown that in equimolar, asymmetric lipid bilayers, the two leaflets of the bilayer are in tensile and compressive mechanical stress. In response, cholesterol molecules redistribute between the leaflets to relieve these stresses. As a result, the distribution of cholesterol molecules depends on the relative ordering of lipids in the two leaflets. We show that there is a quantitative relationship between cholesterol distribution and the ordering of lipids. We have also studied the distribution of cholesterol molecules in lipid bilayers as a function of temperature and phase of the lipids. We have applied ML to study phase behavior in three-component lipid mixtures and have shown that trained ML methods are quite effective in reproducing the phase diagram of lipids. Furthermore, we have applied ML tools to model the time-dependent corrosion rate of mild steel in presence of corrosion inhibitors injected in different doses and dose schedules. We find that trained ML methods are quite effective in modeling the entire kinetics of the time-dependent corrosion rates.
Sumit Sharma, Associate Professor (Advisor)
Douglas Goetz, Professor (Committee Member)
Gary Weckman, Professor (Committee Member)
Alexander Neiman, Professor (Committee Member)
Marc Singer, Associate Professor (Committee Member)
159 p.
Biomedical Engineering; Chemical Engineering
Molecular simulation; Machine learning; Cell membrane; Lipid bilayer; Lipid; cholesterol; Phase diagram; Martini force field; Corrosion inhibitor; Corrosion rate

Recommended Citations

Citations

  • Aghaaminiha, M. (2021). Application of Molecular Simulations and Machine Learning Methods to Study Biological and Metallic Interfaces in Aqueous Environment. [Doctoral dissertation, Ohio University]. OhioLINK Electronic Theses and Dissertations Center. http://rave.ohiolink.edu/etdc/view?acc_num=ohiou162818080313617

    APA Style (7th edition)

  • Aghaaminiha, Mohammadreza. Application of Molecular Simulations and Machine Learning Methods to Study Biological and Metallic Interfaces in Aqueous Environment. 2021. Ohio University, Doctoral dissertation. OhioLINK Electronic Theses and Dissertations Center, http://rave.ohiolink.edu/etdc/view?acc_num=ohiou162818080313617.

    MLA Style (8th edition)

  • Aghaaminiha, Mohammadreza. "Application of Molecular Simulations and Machine Learning Methods to Study Biological and Metallic Interfaces in Aqueous Environment." Doctoral dissertation, Ohio University, 2021. http://rave.ohiolink.edu/etdc/view?acc_num=ohiou162818080313617

    Chicago Manual of Style (17th edition)

ohiou162818080313617
224
© 2021, all rights reserved.
This open access ETD is published by Ohio University and OhioLINK.