I can't possibly spend the time on this new study that it deserves, but I really am going to think about it on my commute today. Or listen to the Halo Effect album I didn't know about until yesterday because that's how busy my 2025 has been. And maybe also think about this paper.
When Anna Barker was on THE Proteomics Show podcast she stressed how absolutely critical the sample handling was to the setup of CPTAC (that was pretty much her idea, btw). I asked her what she thought about all the people who are just pulling from repositories and doing studies on historical material and, best I can recall, she wasn't optimistic about the value of those results.
I don't feel like this study is either..... In a big hospital clinic like the one I worked in for years, blood would come in from upstairs really fast, and then we'd have these big drops of blood daily from remote clinics. Some would arrive on ice for specific assays but most would arrive room temperature. Is the proteome of the dude upstairs the same as his identical twin who had his blood pulled at the clinic a half hour away but didn't arrive at the main hospital for 4 hours? When that blood is deposited in a huge biobank, is that data conserved? Maybe now it is? I'd be confident betting that our IBM XTs (not kidding) did not have the capacity in their databases to retain transfer time information, particularly if it was coming in for an assay where it didn't matter.
Stuff we could totally handle, if we knew that it was important. This study suggests that it definitely could be.
No comments:
Post a Comment