Loads of proteins work by cutting other proteins and splicing them together into new forms with altered or completely new functions. If that's what your collaborators are actually interested in, global proteomics can be a pain in the butt. I've had a PC just about bricked for 5 days processing 170 TIMSTOF files with DIA-NN and found out during the data review that my spectral library didn't contain the single peptide that my friends actually cared about. Considering that relative costs per sample of the $1.3M list price TIMSTOF Flex + MALDI 2 that I used for 2 weeks of data acquisition vs. our very nice, and definitely not $1.3M list price Agilent Ultivo (the world's largest HPLC isn't nearly as large in person as it appears in the ads). You might argue successfully that had everyone been using the exact same language, I could have very easily targeted the single peptide variant...
...to be fair, however, finding the sequence variant was NOT FUN. It was not listed in UniProt or NCBI as a peptide isoform variant, so I did have to find the nucleotides and work out the codons and make my target peptide, and send it to PROSIT to make my targeted list (I am specifically just reprocessing the diaPASEF data for that variant now). Skyline LOVES 170 diaPASEF files, btw. I expect to have quan well before US HUPO.
And this is why I liked this new protocol. There are multiple tools listed in the review for converting DNA sequences to splice variants and super easy instructions (and pitfalls) for targeting these in Skyline. Given that these authors say that OVER 90%!?!?!?!? of human genes go through some sort of splicing!?!??!?! I think this protocol might go into the permanent folder on my desktop that just says SUPER USEFU (maybe if it was lower case the L would fit, but I think I need the caps here).
No comments:
Post a Comment