CCL: AutoDock3 and shuffled peptide sequences

 Sent to CCL by: "Dr. Csaba Hetenyi" [csaba[#]ovrisc.mdche.u-szeged.hu]
 Hi,
 For larger peptides larger number of energy evaluations and population
 size may be necessary. For validation of the parameters of LGA it is
 necessary to measure RMSD between the X-ray position and the docked
 peptide. Grid spacing of .3 A helps in some cases. You may trust your
 results and actual LGA (grid) parameter setup after validation.
 Best,
 Csaba
 On Sat, 12 Nov 2005, Starr Hazard hazards+*+musc.edu wrote:
 > Sent to CCL by: Starr Hazard [hazards|-|musc.edu]
 > Folks,
 >
 > I am examining the docking of a hexapeptide to a protein with AutoDock3. I
 > generated some initial results and sought to examine specificity. I
 > shuffled the sequence (ie left amino acid content constant and altered the
 > sequence order) and tried to AutoDock3 again. In many cases the shuffled
 > sequences show better scores than the original sequences.
 >
 > The shuffled and "wildtype" sequences were docked under similar
 ie mostly
 > default conditions. Have I shown that the binding is none specific or have
 > just got a sampling error?
 >
 > Starr
 >
 >
 > Starr>
 >
 >
 >
 ------
 Csaba Hetenyi, MSc, PhD
 Dept. of Medical Chemistry, University of Szeged,
 8 Dom ter, Szeged 6720, Hungary