On June 15, 2024, an easy-to-use Windows Desktop Application version of the PRAS server was developed. The download link for the EXE file is available through the website https://www.protein-science.com/
While the web application interface performs only the tasks described in the PRAS paper, the desktop counterpart, which I built in 2024 has extended features. The extended features include tools to build model structures (such as constructing a model of a protein in a pure solvent or mixed solvent, generating pure or mixed solutions, etc.), tools to build crystalline nanostructures (spherical or cylindrical silica nanoparticles) and tools to perform Brownian Dynamics (BD) Simulations of colloidal gels.
The program has been tested on Windows 10 and 11 64-bit operating systems. The program follows a standard Windows installation process. To install the software, double-click the installation file. A shortcut is automatically created on the desktop upon completion. To uninstall the software, navigate to Control Panel -> Programs and Features, then double-click to uninstall.
It should be stated that this desktop app integrates the fast and accurate sidechain packer (FASPR) program from which it obtains a flexible dihedral angle (chi) if needed to fix the missing heavy atoms. The app will only warn the user that it is using the default chi for the repair when an mmCIF file that FSPR cannot read is uploaded. This behavior is consistent with the online server. The user should refer to the PRAS paper (https://pubs.acs.org/doi/10.1021/acs.jcim.2c00571) to understand why PRAS uses chi generated by FASPR for protein repair. Therefore, the desktop app generates the same result as the online PRAS server. The user can investigate this and if there are differences, do not hesitate to contact me.
The six radio buttons seen in the figure below are unchecked and represent the default behavior. Therefore, to add hydrogen atoms is TRUE, to assign secondary structure elements (SSE) is TRUE, to plot four Ramachandran types is TRUE, to keep ligands is FALSE, to consider Histidine protonation at the pH of 7 is FALSE (HIS is neutral in this case), to retain the temperature factor is FALSE, to use atoms with the highest occupancy is TRUE and the chain to process is ALL. The user is required to set the working directory. If the PDB structure file to be repaired is stored locally on the user’s computer, it must be located in the working directory. The user must have permission to write in the working directory. For this reason, it is suggested to use common directories like the Downloads or Desktop folder. When using the PDB ID, the program downloads the file automatically from the protein data bank and stores it in the user’s working directory before proceeding with the repair (the user must be connected to the internet when using the PDB ID).
Initially, this tool was designed to generate randomly placed particles in a box for BD simulations. However, with the intention to implement Langevin dynamics, Molecular dynamics, and Molecular docking tools, I expanded its functionality to generate various model systems that I may study or simulate in the future. Currently, the tool allows users to build model systems with up to three molecule types. Examples of model systems generated by this tool is shown in the figure below. A user manual detailing the usage is included with the distribution
To build a nanostructure, the user should first select the desired shape. For a spherical SiNP, only the diameter is required. The user should enter the diameter in the provided field. The radio button labelled “Do not functionalize” is unchecked by default, meaning that to functionalize the SiNP is TRUE. The functionalization employed here is simply to cap the dangling bonds with hydrogen atoms in order to maintain neutrality. Si atoms are satisfied by 4 covalent bonds, and O atoms by 2. However, when the structure is generated, the ends of the SiNP will have unsatisfied Si and O atoms, resulting in a charged overall structure. If this is the user’s expected output, the radio button should be checked. The user should then click the “APPLY” button to generate the SiNP. For a cylindrical SiNP, the user should enter the diameter and length in the provided fields. The diameter must be less than the length. All dimensions are in angstroms. The figure below shows nanostructures generated by the app.
I implemented Brownian dynamics (BD) simulation for the study of the viscoelastic behaviour of colloidal gels. Each colloidal particle represents a protein or carbohydrate molecule while being aware of the limitations of this model, namely, proteins have bonded, angle and dihedral interactions and exhibit pH-dependent charge heterogeneity. Here, the colloidal gels are prepared under equilibrium condition where natural fluctuations govern the process. The user should follow the tutorial in the user manual to simulate the gel as shown in the figure below. The BD simulation is conducted in reduced units as often the case in the literature. The energy shown in the figure below is the total energy and not the energy per particle. The user can divide the total energy by the number of particles to obtain the energy per particle.
This test is used to measure the viscosity of the gel at various shear rates, using Lees-Edwards boundary condition. This is a non-equilibrium steady state simulation. A user manual detailing the tutorial is included with the distribution. The viscosity of the gel obtained at various shear rates is shown in the figure below.
This test is used to measure the viscoelasticity of the gel by computing the storage and loss moduli at different frequencies, using Lees-Edwards boundary condition applied in oscillatory fashion. This is a non-equilibrium steady state simulation. A user manual detailing the procedure is included with the distribution. The storage and loss moduli obtained at various frequencies is shown in the figure below.
This test is used to measure the viscoelasticity of the gel by computing the storage and loss moduli at different strains, using Lees-Edwards boundary condition applied in oscillatory fashion. This is a non-equilibrium steady state simulation. A user manual detailing the procedure is included with the distribution. The storage and loss moduli obtained at various strains is shown in the figure below.
PRAS has been peer reviewed and published. Please cite PRAS as:
O.S. Nnyigide, T.O. Nnyigide, S.G. Lee, K. Hyun. Protein Repair and Analysis Server: A Web Server to Repair PDB Structures, Add Missing Heavy Atoms and Hydrogen Atoms, and Assign Secondary Structures by Amide Interactions. J. Chem. Inf. Model., 2022, 62, 4232–4246.
React
Typescript
Django
Laravel
Facebook
Microsoft
Google
Alibaba
D3
Tencent