Software for identifying co-evolutionary sectors in proteins using "Robust Co-evolutional Analysis (RoCA)"
Co-evolution networks of HIV/HCV are modular with direct association to structure and function
Ahmed A. Quadeer, David Morales-Jimenez, and Matthew R. McKay
-
A PC with MATLAB (preferrably v2017a or later) installed on it with the following additional toolboxes:
- Bioinformatics Toolbox
- Statistics and Machine Learning Toolbox
-
For running codes related to statistical coupling analysis (SCA), register and download the SCA software from https://ais.swmed.edu/rrlabs/register.htm
-
For mapping predicted sector residues on crystal structures, download Pymol available at https://pymol.org/
-
Inferring co-evolutionary networks for a protein using RoCA
- Open MATLAB
- Run the script
main_RoCA.m
and provide the MSA matrix as an input
-
Reproducing results in the paper for HIV and HCV viral proteins
-
Run the following scripts to generate RoCA (and PCA [Quadeer et al. 2014]) results
main_gag.m
for HIV Gagmain_nef.m
for HIV Nefmain_ns34a.m
for HCV NS3-4Amain_ns4b.m
for HCV NS4B
-
Run the following scripts to generate SCA results
main_gag_sca.m
for HIV Gagmain_nef_sca.m
for HIV Nefmain_ns34a_sca.m
for HCV NS3-4Amain_ns4b_sca.m
for HCV NS4B
-
Run the following script (in the GT folder) to compare the performance of RoCA and PCA using binary synthetic data
main_GT.m
-
-
To visualizing the step-by-step procedure and the corresponding output
- Download the html folder
- Open the
main.html
file in your browser
--
[Quadeer et al. 2014] Quadeer AA, Louie RHY, Shekhar K, Chakraborty AK, Hsing I-M, McKay MR. 2014. Statistical linkage analysis of substitutions in patient-derived sequences of genotype 1a hepatitis C virus non-structural protein 3 exposes targets for immunogen design. J. Virol. 88:7628–44. doi:10.1128/JVI.03812-13.
For any questions or comments, please email at ahmedaq@gmail.com.