FunFOLDQA: A Quality Assessment Tool for Protein-Ligand Binding Site Residue Predictions ---------------------------------------------------------------------------------------- Version 1.01 (Dec 2011) (c) Liam J. McGuffin, Maria-Teresa Buenavista, Daniel B. Roche Description ----------- This program can be used to assess the quality ligand binding site residue predictions based on 3D models of proteins. References ---------- Roche, D. B., Buenavista, M. T. & McGuffin, L. J. (2011) FunFOLDQA: A Quality Assessment Tool for Protein-Ligand Binding Site Residue Predictions. PLoS ONE, In press. Installation ------------ No installation is required for the FunFOLDQA program itself after you have downloaded the file. The program is provided in the form of an executable jar file (FunFOLDQA.jar). The instructions below are for Linux (bash) but equivalent commands can be used in other Unix based OSes. Requirements: 1. A recent version of Java (java.com/getjava/). 2. A recent version of PyMOL (www.pymol.org). 3. The TMalign program (http://zhanglab.ccmb.med.umich.edu/TM-align/). Please ensure the TMalign program is working on your system before attempting to run FunFOLDQA. Ensure that you have the correct 32bit/64bit version for your hardware and that the TMalign file is made executable: chmod +x TMalign 4. wget installed system wide. 5. The model quality score for the model used in the prediciton, which can be got from a MQAP program usch as ModFOLD - http://www.reading.ac.uk/bioinf/ModFOLD/ - the code can be run with and without the model quality score Running the program ------------------- You can simply edit the shell script (FunFOLDQA.sh) or you can follow the steps below. 1. (optional) Set the bash environment variable for Java, if you have not installed it system wide, TMalign and PyMOL. e.g. export JAVA_HOME=/home/Liam/jdk1.6.0/ export TMALIGN_HOME=/home/Liam/programs/bin/ export PYMOL_HOME=/usr/bin/ 2. Run the FunFOLDQA examples. For example, using the input files and output directories that are provided in FunFOLDQA_example_input.tar.gz with your own directory configuration (use full paths), example of the output results are provided in FunFOLDQA_example_output.tar.gz. If the path of your model is /home/Liam/FunFOLDQA/T0391_test1/MUProt_TS3, your list of templates was /home/Liam/FunFOLDQA/T0391_test1/T0391_PARENT.dat.nr70.filt, your ligand binding site residue prediction file was /home/Liam/FunFOLDQA/T0391_test1/T0391_1_FN.txt, your target was called T0391: $JAVA_HOME/bin/java -jar FunFOLDQA.jar /home/Liam/FunFOLDQA/T0391_test1/ /home/Liam/FunFOLDQA/T0391_test1/T0391_PARENT.dat.nr70.filt /home/Liam/FunFOLDQA/T0391_test1/T0391_1_FN.txt MUProt_TS3 T0391 Or, if you have java installed system wide: java -jar FunFOLDQA.jar /home/Liam/FunFOLDQA/T0391_test1/ /home/Liam/FunFOLDQA/T0391_test1/T0391_PARENT.dat.nr70.filt /home/Liam/FunFOLDQA/T0391_test1/T0391_1_FN.txt MUProt_TS3 T0391 Or, using the shell script provided: ./runFunFOLDQA.sh java -jar FunFOLDQA.jar /home/Liam/FunFOLDQA/T0391/ /home/Liam/FunFOLDQA/T0391/T0391_PARENT.dat.nr70.filt /home/Liam/FunFOLDQA/T0391/T0391_1_FN.txt MUProt_TS3 T0391 - Optional - to run including a global model quality score from ModFOLD3.0 just append the score to the end of the command e.g. $JAVA_HOME/bin/java -jar FunFOLDQA.jar /home/Liam/FunFOLDQA/T0391_test1/ /home/Liam/FunFOLDQA/T0391_test1/T0391_PARENT.dat.nr70.filt /home/Liam/FunFOLDQA/T0391_test1/T0391_1_FN.txt MUProt_TS3 T0391 0.5998 IMPORTANT: Please also note that you should use FULL PATHS for your input files and output directory, the output directory should also end with a "/" and must contain the input model. Output ------ The predicted BDT and MCC scores are shown in the standard output. An example of the files output for target T0391 can be found in the directory T0391_test_example_output when you unpack the file FunFOLDQA_example_output.tar.gz. The important example files containing information about the prediction are: T0391.pat - Pattern file for the neural network containing the 4 (or 5) predictive feature scores - BDTalign, Identity, Rescaled BLOSUM62, Equivalent Residue Ligand Distance (and Model Quality) T0391.pat.out2 - The FunFOLDQA neural network output file containing the predictive feature scores with the final FunFOLDQA model quality score in the last (8th) column T0470_lig.pdb - PDB file containing a superposition of all ligands and templates with the model Trouble Shooting ---------------- Email me: l.j.mcguffin@reading.ac.uk I will try to respond to your issue as soon as I can! Thanks, Liam