#173 things to add, improve, or otherwise address in XCorr p-value code

[Jeff Howbert]

The following is an accumulated list of things that need to be added, improved, or otherwise addressed in the XCorr p-value code. The order does not imply anything regarding relative importance or priority.

1) static N-terminal mods
2) variable N-terminal and static or variable C-terminal mods
3) set limit on number of mods allowed when calculating p-values
4) assertion to trap p-values that are exactly zero
5) decide on treatment of deltaCn (these currently come out negative)
6) calculation of p-values in Tide-integrated code on Linux is about two-fold slower than standalone code on Windows
7) better heuristics for bounding score range to be considered in dynamic programming
8) vary size of score discretization bins so that similar numbers of bins are used at all precursor masses
9) handle high-resolution MS2 data
10) harmonize handling of precursor exclusion with rest of code
11) make sure default values of bin width and bin offset are uniform and in agreement with rest of code (this may be taken care of by item 12))
12) use Attila's global function for converting mass to bin number
13) in TideSearchApplication.cpp, use iterators instead of traditional C++ loop variables where appropriate (locations are flagged with string '//&&')
14) save best match for each peptide
15) merge crux-pvalue branch into trunk

[William S Noble]

Jeff,
Can you comment on which of these 15 items have been resolved and which ones you consider still needing to be addressed?
Thanks.
Bill