ModelEvaluator: a program to evaluate the quality of a single protein model using 
support vector machines and 1D and 2D structural features

The software is free for academic and scientific use only. All the rights are reserved. A user may download a copy of the software for local use as it is. 
Please don not redistribute the software without Dr. Cheng's written permission.  

Author: 

Dr. Jianlin Cheng
Department of Computer Science
University of Missouri, Columbia
Email: chengji@missouri.edu 

Reference:

Z. Wang, J. Eickholt, J. Cheng. APOLLO: a quality assessment service for single and multiple protien models. Submitted, 2011. 

Z. Wang, A.N. Tegge and J. Cheng. Evaluating the Absolute Quality of a Single
Protein Model Using Support Vector Machines and Structural Features. Proteins,
vol. 75, no. 3, 638-647, 2009. 

------------------------------------------------------------------------------------------
Third-party software used by ModelEvaluator 

ModelEvaluator softwre uses svm_classify program in SVM-light
(svmlight.joachims.org) to classify data points.

Reference to SVM-light:
T. Joachims, 11 in: Making large-Scale SVM Learning Practical. Advances in
Kernel Methods - Support Vector Learning, B. Schölkopf and C. Burges and A.
Smola (ed.), MIT Press, 1999. 

I thank the author of SVM-light (Dr. Thorsten Joachims) for the permit
to include svm_classify in the ModelEvaluator software package

SVM-light is free only for non-commercial use. It must not be distributed
without prior permission of the author. The author is not responsible for
implications from the use of this software.
--------------------------------------------------------------------------------------------

Installation of model evaluator:


Installation of this software is easy. 

1. download or copy model_eva1.0.tar.gz to a place

2. unzip it.

	tar xzf   model_eva1.0.tar.gz

3. open configure.pl, set $install_dir to the dir of model_eva, run configure.pl. 

To use this software, you also need to download and install
1D and 2D structure prediction software: betacon and pspro1.1.
Both betacon.tar.gz and pspro1.1.tar.gz are available at http://casp.rnet.missouri.edu/download.
The installation of betacon and pspro1.1 is easy too. 


----------------------------------------------------------------------------------------------

Test


1. cd test

2. run betacon on a sequence T0288.fasta to generate 1D and 2D features into a directory, assuming betacon has been installed in /home/chengji/betacon/. 

	mkdir output

	/home/chengji/betacon/bin/beta_contact_map.sh T0288.fasta  ./output

3. run model evaluator to evaluate the quality of T0288.pdb model

	../bin/model_eval.sh T0288.fasta ./output  T0288.pdb


	You will get a predicted GDT-TS score: 

	GDT_TS score: ~0.698