CarcinoHPVPred

• This tool can be used to predict the carcinogenic nature of new or already available HPV strains. This tool uses genomic information from two HPV core genes E2 and E6.
• Genomic composition information integrated into ensemle of simple machine learning models such as logistic regression, Support vector machine and kNN.
• Standalone version of CarcinoHPVPred is consist of scripts and models written and tested with perl-5 and python-3.9.9.
• CarcinoHPVPred also available as web-server at http://test5.bicpu.edu.in/CarcinoHPVPred.php. The stand alone version works with linux OS only.

Dependencies:

• Prokka
• perl-5
• python-3.9.9
• sklearn: 1.0.2
• numpy: 1.23.1
• scipy: 1.9.0
• pandas: 1.4.3
• matplotlib: 3.5.2
• joblib: 1.1.0

Installation:

• Unzip the zip file unzip CarcinoHPVPred.zip

• Download and unzip prokka unzip prokka.zip

• Copy prokka directory into CarcinoHPVPred directory

To run program simply type:

cd CarcinoHPVPred/

perl CarcinoHPVPred.pl abc.fasta modelName type_of_input

abc.fasta => genome or predicted genes in fasta format

modelName => logistic_regression or svm or knn or lda

type_of_input => full_genome or E2_E6 or E6