[Tutor] Interactive plots...

David Perlman dperlman at wisc.edu
Thu Mar 27 23:54:49 CET 2008

I am thinking about writing a program which will involve, among other  
things, displaying a plot of a series of numbers.  The idea is that  
you could click on the points and move them to change the numbers.   
Reverse-plotting, I suppose.  It need not be complex; the numbers  
will all be zero or one, and it's only necessary to flip the bits, so  
click-and-drag is seriously overkill.  Really it would be better to  
just double-click on a point to switch it from one value to the other.

Can anyone point me in the right direction?  I have written some  
programs in python before, including TKinter, but this new project is  
beyond the point that I know where to even start looking.  :)

In case you care, the application is in functional brain imaging; the  
brain scans generate a certain number of time points (say 500) and  
then the motion of the subject is also calculated.  Standard practice  
is to generate a "censor" file composed of zeros and ones, where zero  
indicates that that time point had excessive motion and must be  
disregarded.  I want to display a graph of the motion over time, and  
allow quick and easy interactive editing of the censor time series in  
visual parallel to the motion graph.  This would save a lot of time;  
at present everyone does this in Excel, which being a horrible  
Windows program can't be integrated into the predominantly UNIX-based  
processing pipeline.  And in any case, it requires manually typing  
all the zeros, looking back and forth between the graph of motion and  
the list of numbers.

I have already written a program to algorithmically generate the  
censor time series from the motion data, but it is absolutely  
essential to be able to manually double-check and if necessary make  
minor edits.  I'd like to be able to keep that functionality in  
Python rather than sending everyone back to Excel...  if possible!

Thanks very much for any help.

"Pseudo-colored pictures of a person's brain lighting up are
undoubtedly more persuasive than a pattern of squiggles produced by a
polygraph.  That could be a big problem if the goal is to get to the
truth."  -Dr. Steven Hyman, Harvard

More information about the Tutor mailing list