Dennis, I'm interested in following your progress on this. Something that would help me would be to make your links more useful by displaying a meaningful name rather than the document URL. I've changed one as an illustration.
To do this in the WIZWIG editor, type the the text for what you want the user to see and then select it. Next click on the "chain link" icon (top row of the editor tool bar). You don't need to fill in all the options; the important one is the link URL.
Incorporating G-quadraplex prediction
I recently discovered that RNAFold now supports G-quadraplex prediction, but it needs the command line switch -g or --gquad to turn it on. This would very useful for the current round's Impossible Folding II - The Intramolecular G-quadruplex lab.
Current release of Vienna package
This is petty of me to mention, but the ensemble frequency the 2.1.1 version gave my T-thermophilise design "Indian Corn" was only 45.5%, compared to the 48.1% I had very carefully crafted using the current web server version. :-( . It looks like the most recent version is 2.1.3, and apparently it is an improvement over 2.1.1. :-)
(Please don't feel compelled to rerun everything for this!)
RE: Current Release of Vienna Package.
I am currently on travel (through the rest of October), but when I get back home, I plan to try to figure out what has changed in Vienna 2.1.3 versus 2.1.1 and update my scripts and executables. The script for analyzing current lab releases will be easy to update, I think, but the script for analyzing past labs has a statistical forecasting module that will be tricky to update...
Thanks for the update, Dennis. But I'm sincere when I say please don't do it just for me. I suspect you have bigger fish to fry.
Image format conversion
Recently, I also needed to convert those PostScript files to bitmaps, in order to feed them to ffmpeg and produce animations like the one you can see on Conformational 6. I smply used ImageMagick's convert tool, wrapped in a bash script. Possibly, you could do something similar by running this tool (you're already running RNAfold instances, I don't suppose that you're using Python bindings directly into the ViennaRNA library), or if you want, you could as well try to use one of the Python interfaces of ImageMagick.
A warning though: as you know, the dot plots generated by RNAfold differ from the ones we're used to in EteRNA. If you want to produce dot plots that look like our EteRNA ones, you will either need to patch RNAfold, or to edit the PS files before converting them to PNG or JPEG. I can help, but if you're planning to use the outputs as they are, you can just ignore this last paragraph :)
Bound pair frequency as a prediction factor
I would be interested in seeing whether the percentage of bound pairs (CG, AU and GU) has value as a predictor. I certainly believe it will, if only because it will be correlated with the Berex predictor. But I would think it hold some additional predictive value.