Toxinpred2.0

Introduction

ToxinPred2.0 is developed for predicting, mapping and scanning toxic proteins. More information on ToxinPred2.0 is available from its web server http://webs.iiitd.edu.in/raghava/toxinpred2. This page provide information about standalone version of ToxinPred2.0.

Installation

To install the package, type the following command:

pip install toxinpred2

Minimum USAGE

To know about the available option for the package, type the following command:

toxinpred2.py -h

To run the example, type the following command:

toxinpred2.py -i peptide.fa

Full Usage:

Following is complete list of all options, you may get these options by "toxinpred2.py -h" 

usage: toxinpred2.py [-h] -i INPUT [-o OUTPUT] [-t THRESHOLD] [-m {1,2}] [-d {1,2}]

Please provide following arguments

optional arguments:
  -h, --help            show this help message and exit
  -i INPUT, --input INPUT
                        Input: protein or peptide sequence in FASTA format or
                        single sequence per line in single letter code
  -o OUTPUT, --output OUTPUT
                        Output: File for saving results by default outfile.csv
  -t THRESHOLD, --threshold THRESHOLD
                        Threshold: Value between 0 to 1 by default 0.6
  -m {1,2}, -- model Model
                        Model: 1: AAC based RF, 2: Hybrid, by default 1
  -d {1,2}, --display {1,2}
                        Display: 1:Toxin peptide, 2: All peptides, by
                        default 1

Input File: It allow users to provide input in two format; i) FASTA format (standard) (e.g. peptide.fa) and ii) Simple Format. In case of simple format, file should have one peptide sequence in a single line in single letter code (eg. peptide.seq).

Output File: Program will save result in CSV format, in case user do not provide output file name, it will be stored in outfile.csv.

Threshold: User should provide threshold between 0 and 1, please note score is proportional to toxic potential of peptide.

Models: In this program, two models have been incorporated; i) Model1 for predicting given input peptide/protein sequence as toxic and non-toxic peptide/proteins using Random Forest based on amino-acid composition of the peptide/proteins;

ii) Model2 for predicting given input peptide/protein sequence as toxic and non-toxic peptide/proteins using Hybrid approach, which is the ensemble of Random Forest+ BLAST+ MERCI. It combines the scores generated from machine learning (RF), MERCI, and BLAST as Hybrid Score, and the prediction is based on Hybrid Score.