A future without global wiggle
| Case number: | 699969-990039 |
| Topic: | General |
| Opened by: | Seagat2011 |
| Status: | Open |
| Assigned: | Anonymous |
| Priority: | 3 |
| Type: | Suggestion |
| Opened on: | Friday, July 22, 2011 - 17:16 |
| Last modified: | Saturday, July 23, 2011 - 22:05 |
If rebuilds are not random, and in fact are pulled from a datatbase, then we need to derive an equation, perhaps like one I mentioned earlier, that can normalize or even avergae (the score) of rebuilds having similar shapes, as a way to optimize out wiggle backbone and global wiggle. This will make work on puzzles even faster.
Sat, 07/23/2011 - 22:05
#2
ah, so there really is no benefit to merging them, because there is no way for the fragment insertion to undergo a hidden markov-based analysis based on the codons.
![](http://128.95.1.228:8000/tracking_pixel.gif?events=scAdd&f=1111&wdata={"x":12, "y":11}){kind=tracking}
Interesting idea, but you might want to know that the two tools are different from a technical standpoint:
* Global wiggling are gradient-based optimization;
* While rebuilds are fragment insertions from known structures.
Hence, applying the strategy used for rebuilds for wiggling might actually be even LESS effective.