SIDEpro2.0
Copyright (c) 2013 Ken Nagata, Arlo Randall, & Pierre Baldi
Institute for Genomics and Bioinformatics
University of California, Irvine
http://downloads.igb.uci.edu/

SIDEpro predicts side-chain conformations for protein backbones. The software can be 
applied to proteins of any size, to large protein-only complexes, and proteins complexed
with non-proteins (DNA, RNA, ligands, etc).

The software is free for academic use.
To obtain a license for other purposes, please contact: igb-license@ics.uci.edu

********************************************************************************
INSTALLATION INSTRUCTIONS

Download the file: sidepro.tar.gz and open it in the directory where you want to run it.
The following files should appear:

SIDEpro		    		this is the executable
rotamer_library4.txt		rotamer library file
2AHP.pdb			pdb example
2AHP.sequence.fixed3.txt	example of sequence file to fix first three amino acids.
2AHP.sequence.phosphorylate.txt	example of sequence file to phosphorylate tyrosine
AZK.pdb                         AZK residue in 2AHP.pdb.
cosmos_sidepro                  COSMOS 

*You can copy or move the executable to another directory, but the rotamer library file
must also be copied/moved to the same directory as the executable.

********************************************************************************

USAGE

-i <input_file.pdb>

   The only required input is a file which contains a protein backbone in PDB format.
   By default the output file with the predicted sidechains is named: input_file.SIDEpro.pdb  
      
   Side-chain atom coordinates will be predicted for each residue with coordinates for
   the backbone atoms N, CA, and C. These are the only atoms read in by SIDEpro, all 
   other atoms and all non-atom lines in the file are ignored by default.

   In addition to natural 20 amino acids, SIDEpro will predict sidechains of the following
   amino acids.

   ABA	ALPHA-AMINOBUTYRIC ACID 
   CSO	S-HYDROXYCYSTEINE 
   CSD	3-SULFINOALANINE
   CME	S,S-(2-HYDROXYETHYL) THIOCYSTEINE
   OCS	CYSTEINESULFONIC ACID
   KCX	LYSINE NZ-CARBOXYLIC ACID
   LLP	2-LYSINE(3-HYDROXY-2-METHYL-5-PHOSPHONOOXYMETHYL-PYRIDIN-4-YLMETHANE)
   MLY	N-DIMETHYL-LYSINE
   M3L	N-TRIMETHYLLYSINE
   MSE	SELENOMETHIONINE
   PCA	PYROGLUTAMIC ACID
   HYP	4-HYDROXYPROLINE
   SEP	PHOSPHOSERINE, Phosphorylated serine
   TPO	PHOSPHOTHREONINE, Phosphorylated threonine
   PTR	O-PHOSPHOTYROSINE, Phosphorylated tyrosine


OPTIONS

-o <output_file.pdb>

   Use this option to change the name of the output file. A full path can be provided
   to place the file in a directory other than the current directory.
   
-f <fixed_file.pdb>

   SIDEpro treats all atoms in this file with ATOM or HETATOM records as fixed when 
   making the prediction on the input file. 
   
   You can only provide one fixed atom file; however, you can append multiple files
   together and submit the combined file if needed.
 
-n <non_standard_amino_acid_1.pdb> <non_standard_amino_acid_2.pdb> ...

   Use this option to provide files containing one or more non-standard amino acids(NSAs). The mulitiple
   rotamers are defined by providing multiple pdb files. The rotamer probabilities for each rotamer
   will be set to equal values. The residue index and chain identifier of this file will be matched 
   to the residue in the required input pdb file, given by option -i. The NSAs will replace the native 
   residue, and the chi 1 angle and rotamer will be set to minimize energy between the NSA and the rest 
   of the structure.
   
-s <sequence file>

   Use this option to provide a file that defines which residues are fixed and which are searched.
   Upper-case letters specify that the residue should be predicted and lower-case indicates that its 
   should be fixed to those in the supplied PDB file. Note that upper-case residues can be set to 
   any residue type (for building homology models, etc), whereas fixed residues (lower-case) must 
   match the residue type in the PDB file. Three letter codes are valid if they are surrounded by
   squre brackets (ex. [PHE]). To specify the 15 frequenct amino acids, three letter codes are only
   available.
   
   The sequence should be provided on the first line of text without a header (not fasta format).
   The correspondence between residues in the sequence file and the PDB file is simply the ordering.
   The indicies of the residues in the PDB file are ignored. All text after the first line is ignored.
  
   The followings are valid sequences for PDB file which size is four and sequence is RMKQ.

   1) RmKQ
   2) [ARG][met][LYS][GLN]
   3) Rm[LYS]Q
   4) RM[PTR][SEP]

   2AHP.sequence.fixed3.txt and 2AHP.sequence.phosphorylate.txt are examples of sequence file for 2AHP.pdb.
   
EXAMPLES:
  1) The following is the command to run SIDEpro.

     ./SIDEpro -i 2AHP.pdb

  The predicted sidechain is saved in 2AHP.SIDEpro.pdb. You can specify the output file name by the 
  following command.

     ./SIDEpro -i 2AHP.pdb -o 2AHP.SIDEpro.pdb

  Since the input contains non-standard amino acid and this command did not specify NSA type, 14th
  residue will be predicted as GLY instead of "AZK" type.  

  2) To predict sidechain of AZK type for 2AHP, you need to predict structures of AZK using COSMOS.

     python ./cosmos_sidepro/cosmos_sidepro/cosmos_sidepro.pyc AZK.pdb AZK/

  Use python 2.7.1 for this command otherwise COSMOS will not run. OEChem Tookit and OEChem Szybki are
  needed to be installed to run COSMOS(read cosmos_sidepro/README.txt). The predicted strucures are saved
  under AZK/ folder with the the command above.

  Then run SIDEpro with the following command.

     ./SIDEpro -i 2AHP.pdb -n AZK/AZK_0.cosmos.pdb  AZK/AZK_1.cosmos.pdb  AZK/AZK_2.cosmos.pdb  AZK/AZK_3.cosmos.pdb  AZK/AZK_4.cosmos.pdb  AZK/AZK_5.cosmos.pdb  AZK/AZK_6.cosmos.pdb  AZK/AZK_7.cosmos.pdb  AZK/AZK_8.cosmos.pdb  AZK/AZK_9.cosmos.pdb