You know how we went from
"DIA is cool, but WTF do we do with all of these crappy noisy spectra??"
"We have 14 tools that do amazing neural learning machine stuff that we trust so much that we won't even look at these crappy noisy spectra anymore, they're probably okay cause this machine told us it was"
(That was a fun ABRF core director meeting when everyone was like....ummmm....what....guess this is what we do now...???)
That stuff is probably all now very well validated, or if it isn't, it will be very hard to find anyone to blame for it now. Point of fact, though, deep learning intelligence thingies are much much better now. You can even make them write code for you!
What if you applied this knowledge to D D A data? Well, that's what Scribe is!