What does protein quan mean in bottom up proteomics?

 

Need whacked over the head with some reality today? Me neither! However, this is really worth taking a look at....

I'm reasonably sure that the title is a somewhat-less-than-subtle message, that gets somewhat less subtle as you read through it. 

One main point is that if you've got a protein with a ton of coverage (they use a CPTAC study and a really cool virus Calu-3 culture dataset I think a lot of us are familiar with) assembling peptide quan back into protein quan works great. That is, unless this is a protein that has a ton of different proteoforms, then things get a lot less fun. 

Amyloid beta precursor protein has been linked to a bunch of neurological (and other?) diseases and these authors point out at least 20 different proteoforms described to date. This gene product is cut and modified and modified and cut and modified again and the resulting proteoforms aren't just random gibberish. They're all important. When you take a target like this and digest it with trypsin and then start reassembling your results into a "majority protein ID" or "protein group" you might have just thrown out all of the cool data. 

What these authors show? Yeah...it isn't a "might" it's a "you just threw out anything about this protein that might help you."

What's here besides buzzkill? Demonstrations that top down proteomics, even with the current protein size limitations it has today, can help here. Beyond that, I think that it is a reminder that just because you didn't find anything quantitatively different at the protein group level that doesn't mean you just wasted months. Sometimes (increasingly often?) you need to really dig into your data and try to find the patterns here. The good thing, though, is that LCMS based shotgun proteomics has never been great at identifying or quantifying proteins. It's good at peptides. 

As an aside, I saw a really good talk by Judith Steen recently. I think it was at ASMS 2021 and she had a very similar message for the Tau(?) peptides her group studies. Maybe that means if you're doing neurological stuff you should page extra special attention at the peptide level? Or maybe this is just the best examples. Either way, plenty of work left to do out there!