Final Post

Due to health problems, my work on MTBBASE cannot be continued. The database is however maintained now by UniProt-GOA who already distributed it before. I'm sure MTBBASE is in the best hands there! This also ends this blog, but you can follow my general activities on Google+.

MTB-GOA version 19

It's time for the yearly MTB-GOA release which adds about a hundred papers from 2010 but also some missing earlier ones. Due to some proteomics papers the number of covered proteins increased to 2,455 which covers 61 per cent of the M.tb. genome and a much higher percentage of the actually expressed proteome. We are now at 7,739 annotations.

MTB-GOA, as part of the MTBBASE effort to make accessible the physiology of Mycobacterium tuberculosis, will continue to be released at least in a yearly frequency.

MTB-GOA version 18

The first release in the new year brings a few requested corrections and more F->P annotations. As to the latter, it's at the moment a bit fuzzy how I will proceed with them. There seem to be no errors---at least I hope I would hear of them---but still, I will seek some feedback before I continue with F->P. Due to other work, this may well be the last release before the major one in July. Until then!

MTBBASE on github

I have put all MTBBASE code and the DB on github, so you can not only benefit from the code but also contact me with issues about the database, just like you would do with any other piece of open source software. You can also download the annotations from there. Isn't github awesome?

MTB-GOA version 17

Nine missing papers and a few corrections in this quick release. It looks we were wrong with the prognosis all automatically findable cases of missing process annotations are done, so expect some more of these. Fortunately on the other hand, the list of missing papers shrinks fast, and up to 2009 there seem to remain less than ten missing papers for annotation. Cheers!

MTB-GOA version 16

As promised, this update contains the other half of the "processes for every activity" fix. It also has a few other fixes, additions and missing papers, so it's worth updating otherwise as well. This is probably the last big semiautomatic fix, and maybe the last big error fix for this database. From now on fixes will likely be done by paper review, which is a slow process. The other fixes to be done are frequent specialization of annotations, esp. whenever the GO hierarchy changes. This also means we now have time to expand the database scope---about which details in a later post!

MTB-GOA version 15

Sorry for not announcing version 14 with some missing papers in the meantime. Anyway, 15 brings not only process annotations for over 50 proteins that had none, but also several error corrections of early work. Expect more in both lines next time.