On 21/12/2012 16:26, Vasudevan, Geetha wrote:
I have been trying to figure out the origin for the difference in computed pI
value between EMBOSS-IEP (or EMBOSS-PEPSTATS) and Vector NTI (which uses
Expasy Compute pI/MW).
I get 7.07 with EMBOSS and 6.65 with Expasy. I also tried the web site of
Expasy and I get the same as VectorNTI.
Do you know the origin of the variation?
We use amino acid data values calculated for native amino acids. Expasy
uses values calculated from protein isoelectric focussing (e.g. for 2D
gels) under specific conditions, with a note that for highly basic
proteins these values may not approximate experimentally determined values.
There are multiple possible data values that can be used for isoelectric
point calculations, for example Wikipedia gives a third set without
attribution.
You can choose your own pKa values for each amino acid by using a local
copy of file Epk.dat.
For the next release we will add further documentation on the origin of
the default values used by EMBOSS, and may add a choice of pKa data sets
depending on the reason for the pI calculation and make the Epk.dat file
name specifiable by a new command line qualifier.
Hope this helps,
Peter Rice
EMBOSS Team
_______________________________________________
EMBOSS mailing list
EMBOSS@lists.open-bio.org
http://lists.open-bio.org/mailman/listinfo/emboss
