From the Bax Group at the National Institutes of Health ...
Proline Omega angle prediction from sequence and chemical shifts
As described in the paper:
See Also the PROMEGA Prediction Server Web Site:
RedHat Linux (Fedora Core)/Mac/Win32 version (v 1.01, last updated Oct 16, 2009)
The download archive can be unpacked in unix by:
tar -zxvf promega.tar.Zor with traditional Windows zip software.
The shell script "install.com" in the package can be used to set up the program.
Promega prediction from NMR chemical shifts has been tested using a newly constructed database, from which the chemical shift patterns of the proline and the neighboring residues were derived. When Pro CG chemical shift is available, along with CB and other backbone chemical shifts of the Pro and its neighboring residues, Promega prediction provides an excellent indicator for the Xaa-Pro dipeptide bond, and the Xaa-Pro cis or trans state will be identified unambiguously for the vast majority of cases. When the Pro CG chemical shift is not available, Promega prediction still provides a strong indicator for most Xaa-Pro dipeptide bonds in proteins.
When no chemical shifts are available, the ability of Promega to identify cis Xaa-Pro peptide bonds is much decreased but the likelihood for cis-Pro calculated by Promega still provides a useful indicator for the likelihood that the Pro residue is in the cis configuration.
Note that Promega only provides a statistical probability for the peptide bond in question to be cis or trans. Its principal utility therefore must be viewed as an alert to exercise caution when building molecular models in the absence of complete sets of NOE data, where the possibility of a cis peptide bond is often ignored.
The Promega program is implemented using C++. The pre-compiled executable file ($PROMEGA_DIR/src/PROMEGA for Linux, $PROMEGA_DIR/src/PROMEGA.exe for Windows, $PROMEGA_DIR/src/PROMEGA.mac for Mac) or the starting script ("$PROMEGA_DIR/promega" for Linux) can be invoked with the "TALOS-like" command-line arguments. A complete list of options can be invoked and generated with a "-help" command-line argument.
Use of Promega requires definition of an environment variable "PROMEGA_DIR" or one command-line argument "-promegaDir" to specify the Promega installation directory. If the starting script ("$PROMEGA_DIR/promega" in Linux/Mac) is used, this variable is pre-defined before running the program.
Other files of the Promega package include:
Use of Promega to predict backbone chemical shifts involves the following steps:
VARS RESID RESNAME_PREV RESNAME RESNAME_NEXT CS_CNT1 CS_CNT2 CS_CNT3 CG_SHIFT P_CIS P_CIS_NORM FORMAT %2d %1s %1s %1s %2d %2d %2d %8.3f %8.3f %8.3f 49 V P V 5 3 5 25.500 1.000 0.998 54 D P M 5 3 5 28.300 0.000 0.000 62 W P F 5 0 4 9999.000 0.601 0.074 66 A P R 5 3 5 28.100 0.000 0.000 76 A P H 5 3 5 27.400 0.000 0.000
The program provides two outputs: P_cis (the relative likelihood that the Pro is in the cis form, with P_cis = 0.5 corresponding to the database average of 5%) and the absolute probability, P_cis_norm, that the peptide bond is cis.
The Promega program is written in standard C++. Compiling requires:
cd $PROMEGA_DIR/src make -f Makefile.$YOUR_SYSTEM
Compiling of Promega program has been tested on Linux (Linux 9 or newer), Windows (XP) and Mac OS X 10.5.7. Pre-compiled executable files ("$PROMEGA_DIR/src/PROMEGA" for Linux, "$PROMEGA_DIR/src/PROMEGA.exe" for Windows, "$PROMEGA_DIR/src/PROMEGA.mac" for Mac) are contained in the distributed Promega package.