 |
 |
Hello, So sorry to post this here but I'm at wits end and am just hoping that somebody here has had a similar problem (and subsequent fix). Here goes. I have 7 systems, close to identical, all running FC7. For all, I used the autobuild script on the ccp4 website to generate the entire ccp4 package (including coot0.3.1). As I don't want to deal with network problems, I install the programs on every computer following the same build methods. I know I could do this differently, but for now I don't want to change methods. Now, everything appears to be running normally with no issues. However, I just recently tried building a model from scratch, using the baton-mode and then CAatoms - > mainchain button. One one computer, this works great, and I can build, combine fragments, renumber, blah blah blah and get a structure. For the other 6 machines, I can do the baton mode (and it seems to work), but when I hit the CA - > mainchain button, nothing happens (it just spins for a few seconds then appears to work but there are no new atoms on the screen). I tried downloading the newest ccp4 with no luck, and also the newest coot (from the coot website). They seem to run fine, but again get stalled when I try to do what I state above. It just doesn't work (though coot appears to be working normally - I can display molecules/maps and move things around and such). The problem is, I don't even know where to start here. I would gladly replace/uninstall/reinstall, but what is it that is broken and needs fixing? Like I said, I tried several different reinstalls today with no luck at all (even the new coot gave the same problem). I went to the coot faq and did the soft-link of the reference-libraries (which is an old fix but you never know) and it had no effect. I will say the one machine that is working is my one at home, and so I can't simply share that directory via nfs and see if it works on the other machines (not trivial). All were built the same way, but obviously I missed a file/link or something that is not allowing this important part of the program to work. Any helpful tips? Thanks so much. Dave Roberts [log in to unmask]