<div>Hi All</div><div><br></div><div>I have used g_covar to produce a covariance analysis of a trajectory file and am unsure of the eigenvalues produced.</div><div><br></div><div>I expected the numbers to start small (first six almost zero) and increase as when I used g_nmeig on the Hessian; however, I get the opposite:</div>
<div><br></div><div><div># This file was created</div><div># by the following command:</div><div># g_covar_d -f md.trr -s md.tpr -av average.gro -last 50 -mwa </div><div>#</div><div># g_covar_d is part of G R O M A C S:</div>
<div>#</div><div># S C A M O R G</div><div>#</div><div>@ title "Eigenvalues of the covariance matrix"</div><div>@ xaxis label "Eigenvector index"</div><div>@ yaxis label "(u nm\S2\N)"</div>
<div>@TYPE xy</div><div> 1 414915</div><div> 2 7545.63</div><div> 3 708.598</div><div> 4 190.938</div><div> 5 104.664</div><div> 6 38.1187</div><div> 7 34.4296</div>
<div> 8 28.4192</div><div> 9 23.6477</div><div> 10 20.7779</div><div> 11 17.4987</div><div> 12 15.2986</div><div> 13 14.9555</div></div><div> etc.</div><div><br></div><div>
I selected 'not solvent' for both the calculation of the average and the covariance, but the number of eigenvalues produced does not stop at either 50 as I asked for, or after the number of atoms with out solvent. It prints values all the way to the total number of atoms with values of essentially zero after about the first 1000. The eigenvectors does stop at 50 though. </div>
<div><br></div><div>I have searched the archive and can't seem to find any information on this. Does anyone have any experience with this problem? My only thought is to rewrite the .trr and .tpr files with out the solvent and try again, if this is possible?</div>
<div><br></div><div>Thanks for any help,</div><br>Tom<br><br>-----------------------------------------------------------------------------------------------------<br>
Dr. T. L. Rodgers<br><div>Biophysical Sciences Institute</div><div>Durham University</div><div>South Road</div><div>Durham</div>DH1 3LE<br>