You need to setup your X environment. On macOSX hopefully you did this when you first configured the operating system. If you did not, you will want to go to Apple's website to download and install X11.
Another possiblility is that your terminal window has not set the DISPLAY environment variable. Are you running smartnotebook remotely? You may be able to manually fix it with commands like this:
The name of the nmrview executable is probably different than what the installation script thought. No problem, first find out what the name is given to the nmrview executable by looking in your nmrview installation (eg /usr/local/nmrview5.0.4). When you know the name, edit the smartnotebook startup file (eg, /usr/local/snb-v5.x/sbin/snbview) and revise the last line from "/usr/local/nmrview5.0.4/nmrview5.0.4.xxx" to the true location of nmrview.
I have also had some linux machines complain that libtcl.8.3.so is missing. You may be able to copy the shared libraries (libtk.8.3.so as well) from another machine and place them in /usr/lib. Further discussion is also at http://www.nmrview.com/archive/msg00488.html.
Make sure you are executing the nmrview/smartnotebook startup script. A typical location for it is /usr/local/snb-v5.x/sbin/snbview.
Perhaps you have forgotten to include some of the required input files.
Perhaps you made a mistake in manually customizing the init file (init.snb). Either examine and fix your changes or move ./init.snb to ./init.snb.bkp and restart smartnotebook to create a working ./init.snb.
Another possibility is that one of the output files in 'snb.out' may be messed up. Try moving ./snb.out to ./snb.out.bkp and then restart smartnotebook. If it now works, you may want to go to snb.out.bkp and salvage any of your previous work (like the chains file).
Your data directory is empty. Have you downloaded the data files and installed them? See the startup section for details. Also, make sure are in the directory which contains 'hsqc.nv' when starting up smartnotebook.
The current version of smartnotebook has a limit of 2500 picked peaks per file. This is alot of peaks. If you pick too much noise, smartnotebook will get bogged down in attempting to connect all the noise peaks.
Check the referencing of your peakpick files. Use the ref_peakpick script to appropriately alter the peakpick values.
This was a bug in previous versions that should now be corrected. But if it does happen, you can always find an up to date version of the chains in the log file. Find the most recent good version, then cut and paste this text into the file snb.out/chains. Restart snb and the world should be back to normal. Send an email to me that the bug is not fixed.
It's a bug that seems to disappear when I restart smartnotebook. I just haven't figured it out yet.
If the data was processed on a different operating system platform, you may need to toggle the "byteorder" parameter in each par file.
If you have an error that was not covered in this section, then try these things:
All help and latest news is available on-line . At this time we know of no newsgroups or mailing lists related to smartnotebook.
If you think you are going to use smartnotebook alot, then it may be convenient to make an alias based on where smartnotebook was installed and put it in your .cshrc or .bashrc. I am calling the alias "snbview" but call it whatever you like.
alias snbview /usr/local/snb-v5.x/sbin/snbviewRemember the alias takes affect when you bring up another window or source the file.
You can re-install smartnotebook but it is much easier to edit a copy of the snbview script appropriately.
It is just easier for the developer, if you want you can cat all the tcl files into one single file and just source that (especially if the names for my tcl scripts match yours).
The easiest way to do this is to change your font and make it bigger or smaller. The windows will automatically re-adjust their size to fit. You can change your font using the menu item "Options / "Preferences".
One of the smartnotebook examples, called "hsqc_view" shows how this is done. In this example, the font is set to "6x9", which results in very tiny windows.
The peakcon program does not differentiate between pattern types, it scores each pattern based on the worst fitting peak of that particular pattern. What is judged as "worst" is based on the intrafile and interfile errs entered in the rules file. It is easier to find a better fit through a smaller number of peaks than through a larger set.
The dgx pattern requires only one connecting peak joining hncacb(i-1) and hncacb(i) whereas dxg requires two peaks. Given the simple way peakcon calculates connectivities, the presence of two peaks limits the number of false positive connections.
It may be that your referencing is off between peakpick files. I have included a simple script in the bin directory called ref_peakpick that could help to fix the peakpick files.
Likely the chain is assignable to an area you have already assigned. Whether this should be allowed or not is debatable, personally i think it is interesting information. Are you sure you got the first assignment right?
Yes, you don't have to believe smartnotebook, just click on the residue in the sequence panel where you want to assign the chain and then click assign. This can actually happen more often than you think, especially near the end of the process when you want to assign a remaining short chain but it turns out to have a "best fit" at a place that is already assigned. Smartnotebook thinks there is ambiguity but you know better.
This file last updated: Questions to: bionmrwebmaster@biochem.ualberta.ca