<html>
<head>
<style>
.hmmessage P
{
margin:0px;
padding:0px
}
body.hmmessage
{
FONT-SIZE: 10pt;
FONT-FAMILY:Tahoma
}
</style>
</head>
<body class='hmmessage'><div style="text-align: left;">Hi,<br><br>The cosine content can only be use for "negative" conclusions.<br>If it is high, the sampling is bad.<br>If it is low, the sampling can be either good or bad, it doesn't tell.<br></div><br>The overlap is a much better measure.<br>Since it is so low, you have bad sampling.<br><br>Have you looked at the eigenvector inner products to check<br>if the pc directions change, or if only the eigenvalue change?<br><br>Berk<br><br><hr id="stopSpelling">> Date: Thu, 23 Oct 2008 13:10:23 +0200<br>> From: rcsqtc@iiqab.csic.es<br>> To: gmx-users@gromacs.org<br>> Subject: [gmx-users] PCA comparison<br>> <br>> Dear all,<br>> We are performing some PCA analysis of several 22ns trajectories of a<br>> protein hexamer at different temperatures (280, 300, 320K). We expected<br>> to see a similar movement decribed by the ~10 lowest PCA's.<br>> their overlap is very poor: Below 0.35, and the diagonal elements are<br>> not remarkably higher than the rest. If we split one of the trajectories<br>> into to 11ns ones and do the same, again the overlap obtained is below<br>> 0.44 in all cases and there is no clear diagonal. 11ns is quite a long<br>> trajectory so I expected to see the same type of movement described by<br>> both 11ns fragments. Here are the cosine contents of the whole 22 ns<br>> trajectory and the two fragments. Except for the 1st pc, the rest seem<br>> well sampled: shouldn't they agree?<br>> 22ns:<br>> Cosine content PC1 0.956255<br>> Cosine content PC2 0.00428106<br>> Cosine content PC3 0.000162083<br>> <br>> 1st 11ns:<br>> Cosine content PC1 0.923261<br>> Cosine content PC2 0.0217681<br>> Cosine content PC3 0.0135543<br>> <br>> 2nd 11ns:<br>> Cosine content PC1 0.786694<br>> Cosine content PC2 0.1654<br>> Cosine content PC3 0.000335589<br>> <br>> Are we doing something wrong? Is this the exptected behaviour?<br>> This is what we do to calculate the PCA's (10ns of equilibration removed):<br>> g_covar -f trajectory.xtc -b 10000 -e 21000 -s structure.pdb -o<br>> eigenval-T300-A.xvg -v eigenvec-T300-A.trr -av average-pca-T300-A.pdb<br>> g_covar -f trajectory.xtc -b 21020 -s structure.pdb -o<br>> eigenval-T300-B.xvg -v eigenvec-T300-B.trr -av average-pca-T300-B.pdb<br>> <br>> and to compare them:<br>> g_anaeig -inpr T300-A-B.xpm -v eigenvec-T300-A.trr -v2<br>> eigenvec-T300-B.trr -first 1 -last 10<br>> <br>> Thanks for your help and suggestions!<br>> Ramon<br>> _______________________________________________<br>> gmx-users mailing list gmx-users@gromacs.org<br>> http://www.gromacs.org/mailman/listinfo/gmx-users<br>> Please search the archive at http://www.gromacs.org/search before posting!<br>> Please don't post (un)subscribe requests to the list. Use the <br>> www interface or send it to gmx-users-request@gromacs.org.<br>> Can't post? Read http://www.gromacs.org/mailing_lists/users.php<br><br /><hr />Express yourself instantly with MSN Messenger! <a href='http://clk.atdmt.com/AVE/go/onm00200471ave/direct/01/' target='_new'>MSN Messenger</a></body>
</html>