Now we're getting somewhere. My argument against most targeted techniques falls right here: There may be as many (or more) than 1,000,000 variants of proteins expressed in human cells. The most comprehensive studies to-date can only identify a small fraction of these. By targeting, I feel like we're often limiting ourselves to what has already been identified. This is one of my many arguments against the SWATHish techniques where we're looking at matching a MS1 mass found in a total MS range to some potential daughter ions that are found when everything within that same mass range is fragmented. Without a comprehensive library, there are far too many false positives.
In a step toward negating this objection, last month in Nature Methods a paper from Paola Picotti et al., announced a "nearly complete" mapping of the yeast proteome. If you have a complete map, these technologies become a whole lot more feasible in my eyes. You can read more about this paper here, and obtain the map at the peptide atlas here.
Update 4/7/13: You can download this database as a spectral library. Not only that, but you can download it as an Orbitrap/LTQ (High-low experiment) spectral library. This aspect of the database is going into use real soon.
No comments:
Post a Comment