Talk:Task 6: MSUD - Sequence-based mutation analysis
I like your wiki entry. It is short, informative and covers all aspects of this task in a sensible manner. I like how you explained the conclusion you made from the given analysis in your discussion
- sentences you might reconsider: The easiest to spot ... ,Polyphen and SIFT or SNAP respectively both ...
- How did you get the pssm- percentage?
- You maybe could have tried to resolve the homolog mammal issue with a different approach.
- Interestingly snap2 performed by far worst in our case
--Meiera 21:35, 18 June 2012 (UTC)
A few more things. I always sound so negative but let's just say that's owed to the time and amount of energy drinks consumed :)
- Wikipedia is not the nicest reference, but since the amino acid article is locked I'll forgive you ;) Alternatively you could have just cited the Alberts and I would of course have believed you on something so trivial.
- the header of the first table could instead have been a small header and then the rest, a bit more detailled, in a caption
- For the structural mutation figures, I strongly suggest turning of shadows in this case. Esp. while in motion, they are more confusing than helping. The command is 'set ray_shadows,0;'. But apart from that I like that you raytraced them at all :)
- There could have been more discussion of the data, e.g. the first two tables are only shown and not mentioned ever again. Of course the knowledge that can be obtained from this data is limited and I probably overinterpreted things so I have something to write, but if you think it doesn't really tell you anything than I guess that's a result as well
- 'Blosum62-Score is', that is is not supposed to be there is it?
- In Figure 1, there are so few elements in the sets, instead of their numbers you could just show the mutation's names
- I like your discussion and the approach of looking at prediction methods first and only when in doubt consider the rest of the data seems to have turned out as a good idea!