音時雨 ～Regentropfen～

Recently, we're working on making a tree for all functional introns in our database. We have made an ORF alignment including about 500 introns. There are two versions of them, one is longer and have more variable regions, whereas the other one is shorter and basically only have the alignable regions.

So now I'm trying to use the two alignment files to make trees, and then, hopefully we can see difference between these two trees and another quick and inaccurate tree. My method involved using RAxML (I myself don't know much about the algorithm, I only know how to use it...).

Basically, for an accurate tree, 100 times of bootstrap is necessary. It's the first time I used it to deal with such hugh alignment files, but I never expected the 100 bootstrap runs would take that long to run.

I started the program on last Friday, and the shorter file was barely half done, while the longer file was around 37. Then I did some quick math:

For the shorter file, 8000 seconds per bootstrap, and thus, 8000/60/60*100/24=9.25 days.
For the longer file, 12650 seconds per bootstrap, so, 12650/60/60*100/24=14.63 days.
...2 weeks to make a tree,
---That's indeed the tree from hell!!!

I just hope there is no power cut during the two weeks. (・_・;)

#1, Rainbow coloured crayon.

↑ 虹色クレヨン。
↓ 虹色クレヨン？！

Case #2, I saw hymmnos from my alignment

-- Although I know it's grammatically wrong.（´▽｀*）

Finally, capricious weather, it's snowing again.

終わり／The End (・ω・｀)