PDB capsid volume analysis

Table of Contents

• Features
• Usage
• Method
• Installation
• Testing
• Analysis
• References

---

Features

The script virus-particle-structure allows the user to calculate the volume of a viral capsid by parsing a CIF file. The output is:

• PDB entry id;
• Volume in angstroms;

It is also possible to download all of the CIF files that contain information about viral capsids in PDB. This can be achieved by running the get-virus-data script that is found in REST directory. This script performs the following tasks:

• Deletes all the previous CIF files and generated data;
• Retrieves PDB entry ids for viral capsids by using the JSON parameter file and PDB Search API;
• Checks the CIF file integrity using CIF COD Tools [1-3];
• For each PDB virus capsid entry, extracts the genome length from the NCBI database using Entrez Direct [4];

The following picture depicts what can be done with these programs.

---

Usage

Usage itself is simple. Here are some examples:

• To calculate viral capsid volume for a given CIF file:

    ./virus-particle-structure input.cif
    ./virus-particle-structure input.cif > output.tab

• To download all possible viral capsid available in PDB archive (all files are stored in PDB_archive directory):

    ./REST/get-virus-data

• To perform volume calculation for all of the downloaded files in PDB_archive directory:

    ./batch_processing > output.tab

• Or simply, to reproduce the data, that was used in this analysis you can run the following make command:

    make reproduce-data

---

Method

Each CIF file contains atom coordinates only for a part of a viral capsid, a certain subunit. The rest of the atom coordinates that make up the viral capsid must be reproduced using the symmetry operators. The procedure looks like this:

New 3D coordinates are obtained by multiplying the previous ones by a rotation matrix and then applying a translation vector.

Installation

There is no installation needed. Just download the package and the all of the scripts are ready to use.

---

Testing

For testing you can run the following make commands:

• To delete all existing test outputs:

    make clean-tests

• To run tests after deleting the test outputs:

    make test

---

Analysis

After statistical analysis using R, the following results where obtained:

After removing all outliers, the most requent volume of the viral capsids was around 27000000.

After removing all outliers, the most requent genome length for a virus species was around 7500 base pairs.

Next, relationship between these two variables was assesed. By performing shapiro-wilk normal distribution test it was determined, that both datasets were non-normally distributed. Although this was the case, correlation analysis was still performed. The r^2 values was equal to 0.02, which means no linear correlation between these two datasets.

References

1. Vaitkus, A., Merkys, A. & Gražulis, S. (2021). Validation of the Crystallography Open Database using the Crystallographic Information Framework. Journal of Applied Crystallography, 54(2), 661-672. doi: 10.1107/S1600576720016532
2. Merkys, A., Vaitkus, A., Butkus, J., Okulič-Kazarinas, M., Kairys, V. & Gražulis, S. (2016). COD::CIF::Parser: an error-correcting CIF parser for the Perl language. Journal of Applied Crystallography, 49(1), 292–301. doi: 10.1107/s1600576715022396
3. Gražulis, S., Merkys, A., Vaitkus, A. & Okulič-Kazarinas, M. (2015). Computing stoichiometric molec- ular composition from crystal structures. Journal of Applied Crystallography, 48(1), 85-91. doi: 10.1107/s1600576714025904