PSSM Coreceptor Predictor

use predictor:	pre-process:
subtype B X4/R5 subtype B SI/NSI subtype B X4/R5 (Poveda2009) subtype B SI/NSI (Poveda2009) subtype C SI/NSI	Force V3 at position 1 Fast-find V3 Align to matrix Expand degenerate sequence Average score Full expansion

changelog

02 Dec 2009
added explicit sequence type selection control
fixed alignment issue when gaps present in user sequence
fixed sequence expansion bug occurring when V3 is not 
 found on first pass
expanded the warning score limits for X4R5 to [-15, 9.136]
worked around an degen expansion issue
thanks to U Roma 'Tor Vergata'
03 Jun 2009
fixed prediction labels for Pov: thanks Jerry
24 Mar 2009
added choice of Poveda et al, JAC 2009 cutoffs
26 Feb 2009
added 'average only' expansion option
major refactor of code guts
minor bug fixes
19 Feb 2009
now trims to V3 before degenerate expansion
15 Feb 2009
multiple optimal alignments now correctly reported
06 Feb 2009
fixed a bug causing 500 errors when using "Expand"
fixed a bug causing scores/predictions to be wrongly
 reported when using "Expand" (this bug only a problem
 since 03 Feb 2009)
03 Feb 2009
added degenerate sequence expansion (thanks Cathia)
rationalized the pre-processing options
doc updates
04 Jan 2009
implemented 150kb input limit 
(avoids timeouts; just break up your files)
30 Dec 2008
unnamed sequences can be entered in textarea
(separate sequences by a blank line)
24 Dec 2008
fixed scoring bug (thanks Liek & Evelien)
23 Dec 2008
fixed short sequence issue
made alignment and V3 search user-selectable; modified docs accordingly
parried PRH

Fortinbras PSSM