Proteus

• Requires python and the machine learning package 'scikit-learn'
• A slightly modified version of DISOPRED3.0 is distributed with this package
• 'scikit-learn' may require additional packages to be installed
• Also requires the uniref90.fasta database and its associated files in the folder DB
• download the updated-most version of 'uniref90.fasta' file (sequence database) from the web (http://www.ebi.ac.uk/uniprot/database/download.html)
• do a format database on it (formatdb) to generate the associated files
• And empty DB directory is provided with the installation

Installation Notes for scikit-learn

This tutorial requires the following packages:

• Python version 2.6-2.7 or 3.3-3.4
• numpy version 1.5 or later: http://www.numpy.org/
• scipy version 0.10 or later: http://www.scipy.org/
• matplotlib version 1.3 or later: http://matplotlib.org/
• scikit-learn version 0.17: http://scikit-learn.org
• seaborn version 0.5 or later

Installation of scikit-learn in Ubuntu 14.04

sudo apt-get install python-sklearn
sudo apt-get update sudo apt-get install build-essential python-dev python-setuptools python-numpy python-scipy libatlas-dev libatlas3gf-base pip install --user --install-option="--prefix=" -U scikit-learn

Installing Proteus

$ git clone https://github.com/bjornwallner/proteus
$ cd proteus
$ chmod +x proteus/run_proteus.py

The program has just one inputs

    1. A fasta file containing a single amino acid sequence in fasta format

Run Step:

$ ./proteus/run_proteus.py <basename.fasta>

EXAMPLE OUTPUT:

$ cat basename.seq.proteus
#Proteus v1.1
#pos res pred prob
1 M 1 0.518
2 R 1 0.561
3 V 0 0.439
4 K 0 0.416
5 E 0 0.438
6 I 0 0.439
7 R 0 0.392
8 K 0 0.427
9 N 0 0.405
10 Y 0 0.400

and a graphical representation (.png) of the same (Protean segment prediction score vs. Residue)