thanks for your responses to my question regarding the precision of the
technique of peptide mass fingerprinting. I certainly accept that the
technique can be quite accurate in a number of circumstances but I
still feel the approach is being a bit oversold. I would like to restate
my original question in terms of a hypothetical (but I think not
unreasonable) situation to make my concerns clear.
If one derives (enough) sequence from a mystery protein by whatever
technique (Edman or MS) and interogates the database essentially there are
three answers -
1. the protein is known
2. the protein is unknown but, because of sequence similarity, is related
to a known protein, or
3. the protein is unknown
These answers, I believe, come with extremely high levels of certainty.
With a peptide mass fingerprint can these same answers be given and with
what level of confidence?
If you were given, say, 100 spots from a 2D gel of rat liver (sorry, it's
the "simplest" tissue I deal with and I know that the sequence database is
pretty incomplete for the rat but it does reflect pretty common
scenario), what would either of you expect the success rate of peptide
mass fingerprinting to be?
My personal experience (and I would love to know what I am doing wrong if
the problem is in my hands) is that it is this is quite a test for the
technique, especially when you are dealing with proteins in the 100kDa+
range (their theoretical peptide maps look like random number
generators!).
Please don't get me wrong, I'm not trying to pour cold water on the
technique, I'm just looking to understand its limitations, looking forward
to your comments....Ken
********************
Ken. I. Mitchelhill
John Holt Protein Structure Laboratory
St. Vincent's Institute for Medical Research
41 Victoria Parade
Fitzroy 3065 Victoria
Australia
Telephone: 61-3-9288 2480
Fax: 61-3-9416 2676
Email: ken@ariel.ucs.unimelb.edu.au
Webmaster for the Association of Biomolecular Resource Facilities
http://www.medstv.unimelb.edu.au/abrf.html
*********************