On Oct 31, 2007, at 4:09 AM, Bob Hanson wrote: > Timothy Driscoll wrote: > >> hi, >> >> I am using an external data file to apply a roygb color scheme to a >> set of atoms. a snippet from an example data file looks like this: >> snip >> >> my Jmol script loads this file and colors atoms (column 1) by the >> values in the last column (4). the values in column 4 range from -9 >> to 31 for this particular file. >> >> the Jmol output, however, is almost completely yellow and green, with >> a little blue and no red: >> >> <http://www.molvisions.com/temp/colorissue.jpg> >> >> (left panel). this seems odd; I expected a smearing of color. a >> different data file, with values ranging from -3124 to 77334, gives >> an entirely red structure (above url, right panel). >> >> >> my hypothesis is that the range of values is too wide, and maybe this >> would work a lot better if I binned them into 5-10 categories first. >> perhaps even log-transform them. this will require a bit of work, >> though, so I thought I would float it on the list first. can anyone >> confirm this hypothesis? >> > > first confirm that Jmol is treating your data correctly. What is it > reporting for the data range? You could set this range yourself so > as to > allow the outlyers to "overflow" the red/blue range and get a better > distribution around the middle. > yep, the report does match the actual data range, so Jmol appears to be handling the data correctly. I don't know the shape of the distribution, though, and it may change between data sets, so I'm not entirely comfortable imposing range limits. in fact, I didn't know range limits were even possible! (it may come in handy elsewhere, so thanks for the tip in any case. :-)
Bob, can you explain how Jmol assigns colors in the case of an external data set? for example, the roygb scheme. I assume Jmol breaks up the entire data range into 5 equal sections, and applies one color to each section. (for the bwr scheme, it would be three instead of five sections.) is this correct? if so, I can dispense with binning the data and look into transforming them into a normal distribution. cheers, tim -- Timothy Driscoll em: [EMAIL PROTECTED] Virginia Bioinformatics Institute ph: 540-231-3007 Bioinformatics I: M-1 im: molvisions Washington St., Blacksburg, VA 24061 04-16-07. We will not forget you. ------------------------------------------------------------------------- This SF.net email is sponsored by: Splunk Inc. Still grepping through log files to find problems? Stop. Now Search log events and configuration files using AJAX and a browser. Download your FREE copy of Splunk now >> http://get.splunk.com/ _______________________________________________ Jmol-users mailing list Jmol-users@lists.sourceforge.net https://lists.sourceforge.net/lists/listinfo/jmol-users
