Okay. I'm thinking that this week just produced proteomics paper of the year. This is nuts. Deep breaths, Ben...
The paper in question? Joel M. Chick et. al., out of Steve Gygi's lab. Its called: A mass-tolerant database search identifies a large proportion of unassigned spectra in shotgun proteomics as modified peptides.
According to Chick's new paper, I'm just scratching the surface! They went in and addressed the unmatched spectra by using HUGE windows. They opened it up 500 Da. 500Da! I used 10ppm. Even in Byonic searches when I go full out wild-card I only open up the tolerance by 80Da and that takes all day.
What this study comes back with is almost 50% more peptides. They bin the mass discrepancies out and its PTMs. All over the place, PTMs! The FDR is possibly a little high but this study just uncovered what a massive fraction of unidentified MS/MS events are. Holy cow.
Definitely definitely check this paper out. It looks like the answer to SO many questions (like maybe why the Gygi lab is so famous?!?) This is a game changer!
Hi Ben,
ReplyDeleteWe performed these searches on a standard desktop computer. I think we wrote the specs of the computer into the methods. We used a linux compiled Sequest algorithm to do this. To speed up this search it is best to search without dynamic mods and only one missed cleavage event. High resolution MS/MS is ideal to finding these mods. Also, I believe the FDR is accurate. One of the supplementary figures breaks this down.
Regards,
Joel Chick
Joel,
ReplyDeleteAwesome! Thanks for the additional information! I didn't get through the Supp material before I had to head in to work. Phenomenal work here!!!