Phil W
@pwilmarth.bsky.social
Old person forgetting everything they ever knew about proteomics at speeds in excess of 270 Hz...
It must be the start of Land Grab Season! Maybe it is now part of Black Friday deals?
November 29, 2025 at 3:59 PM
It must be the start of Land Grab Season! Maybe it is now part of Black Friday deals?
I am sure it would be better to use XML markdown where the text inside the tags was binhex. The tag labels should all be a controlled vocabulary of acronyms. The schema should guarantee that the resulting file will always be 1% larger than the free space on the storage volume...
November 24, 2025 at 4:55 PM
I am sure it would be better to use XML markdown where the text inside the tags was binhex. The tag labels should all be a controlled vocabulary of acronyms. The schema should guarantee that the resulting file will always be 1% larger than the free space on the storage volume...
4% male means that the young male samples are the noise in the data. That sounds like young males to me.
November 22, 2025 at 3:57 PM
4% male means that the young male samples are the noise in the data. That sounds like young males to me.
Sure, let's give up on the quantitative proteomics technique that works because it is a bit harder than squirting some digest into an autosampler tube. DIA is half crappy data that is not wrong and half wrong data that is even crappier. Single shot studies are noise amplifiers that go to 11. Sigh...
November 18, 2025 at 5:58 PM
Sure, let's give up on the quantitative proteomics technique that works because it is a bit harder than squirting some digest into an autosampler tube. DIA is half crappy data that is not wrong and half wrong data that is even crappier. Single shot studies are noise amplifiers that go to 11. Sigh...
Maybe US HUPO can offer 10% discounts with AARP cards???
November 15, 2025 at 5:36 PM
Maybe US HUPO can offer 10% discounts with AARP cards???
LinkedIck. There is no coming back from "the ick" according to Joanne and Morgan.
November 13, 2025 at 8:30 PM
LinkedIck. There is no coming back from "the ick" according to Joanne and Morgan.
Ha-ha. What YOU want is no longer part of the algorithm.
November 13, 2025 at 8:08 PM
Ha-ha. What YOU want is no longer part of the algorithm.
But if you use multiomics, you can discover bogus biomarkers from any and all of the omics techniques. It is a natural exploration of how many wrongs does it take to make a right...
November 12, 2025 at 11:30 PM
But if you use multiomics, you can discover bogus biomarkers from any and all of the omics techniques. It is a natural exploration of how many wrongs does it take to make a right...
Proteomics has the answers (unlike OLink and SomaLogic): wear your "top down" jacket and try to keep your "bottom up" off of the slippery sidewalks...
November 9, 2025 at 9:06 PM
Proteomics has the answers (unlike OLink and SomaLogic): wear your "top down" jacket and try to keep your "bottom up" off of the slippery sidewalks...
There are more cells in the Mass Spec than in the experiments...
November 6, 2025 at 2:45 PM
There are more cells in the Mass Spec than in the experiments...
Maybe the instructors are confusing science with engineering. What would be beneficial is the high-level scientific gist of mass spectrometry. If you're spending more time on how ions move in electric and magnetic fields than the time it takes to read this sentence, then something is not right.
November 3, 2025 at 6:43 PM
Maybe the instructors are confusing science with engineering. What would be beneficial is the high-level scientific gist of mass spectrometry. If you're spending more time on how ions move in electric and magnetic fields than the time it takes to read this sentence, then something is not right.
Nice presentation, Brett. What to do with the data is such a big hurdle for the entire field. Data analysis and presentation of results is almost always really awful and has not gotten better over time. It is wonderful to see Perseus not being used. Ha-ha.
October 27, 2025 at 11:38 PM
Nice presentation, Brett. What to do with the data is such a big hurdle for the entire field. Data analysis and presentation of results is almost always really awful and has not gotten better over time. It is wonderful to see Perseus not being used. Ha-ha.
If you took all the wasted money spent on these noise measuring platforms, you would have enough money to do something important for mankind like build a giant balls room and gold plate it.
October 25, 2025 at 11:43 PM
If you took all the wasted money spent on these noise measuring platforms, you would have enough money to do something important for mankind like build a giant balls room and gold plate it.
If you're asking about MS1 feature-based DDA, TMT, or (maybe) some DIA where you have relative protein abundances, then check out slides 17-25 in this presentation/blog: github.com/pwilmart/mou....
You'll need more than one normalization step in most experiments. Norms and modeling need to match.
You'll need more than one normalization step in most experiments. Norms and modeling need to match.
GitHub - pwilmart/mouse_lens_development_Khan2018_reanalysis: Re-analysis of Khan 2018 developing mouse lens TMT study (PXD006381)
Re-analysis of Khan 2018 developing mouse lens TMT study (PXD006381) - pwilmart/mouse_lens_development_Khan2018_reanalysis
github.com
October 23, 2025 at 8:27 PM
If you're asking about MS1 feature-based DDA, TMT, or (maybe) some DIA where you have relative protein abundances, then check out slides 17-25 in this presentation/blog: github.com/pwilmart/mou....
You'll need more than one normalization step in most experiments. Norms and modeling need to match.
You'll need more than one normalization step in most experiments. Norms and modeling need to match.