William Beimers

Today my @nytimes.com colleagues and I are launching a new series called Lost Science. We interview US scientists who can no longer discover something new about our world, thanks to this year‘s cuts. Here is my first interview with a scientist who studied bees and fires. Gift link: nyti.ms/3IWXbiE

Congrats Chris!!

Impact of Sample Preparation Strategies on the Quantitative Accuracy of Low-Abundance Serum Proteins in Shotgun Proteomics Serum proteomics plays a crucial role in biomarker discovery and disease research, yet the selection of an optimal sample preparation method remains challenging. Evaluating the accuracy of protein qua...

I'm glad to see more papers that replicate findings that we published earlier this year, but I find it kind of strange that it's almost exactly the same experiments and figures. I'm surprised they avoided citing our manuscript considering it is even in the same journal.

pubs.acs.org/doi/10.1021/...

Congrats to everyone!

[Person with ponytail wearing lab coat hands balloon to another person in front of a springboard, a magnet hanging from above, and then a target] PERSON WITH PONYTAIL to Person 2: Rub this balloon against your head, then go jump past that magnet toward the target on the wall. [caption] Before the bathroom scale was invented, the only way to weigh people was mass spectrometry.

Mass Spec

xkcd.com/3094/

Come check out my Wednesday poster at #ASMS2025 in Baltimore in a couple of weeks for all the information you could want about plasma proteomics methods! My fellow lab member Katie will also be presenting data from a project I am leading on plasma multi-omics of Long COVID. See you there!

We worked hard on this project to get it out quickly hoping that it can serve as a resource for the plasma proteomics community! As plasma studies proliferate and expand, using the right method for each biological question is important. Check out the MS and Olink datasets @pride-ebi.bsky.social.

We have expanded our preliminary findings that were posted to BioRxiv in January, especially when looking at the methods performance in an example cancer study cohort. There are several interesting takeaways, especially comparing Olink to MS methods, that warrant further investigation.

Technical Evaluation of Plasma Proteomics Technologies Plasma proteomics technologies are rapidly evolving and of critical importance to the field of biomedical research. Here, we report a technical evaluation of six notable plasma proteomics technologies...

My first manuscript from @coonlab.bsky.social - Technical Evaluation of Plasma Proteomics Technologies - is now available to read on Journal of Proteome Research! We thoroughly evaluated 5 MS-based and 1 affinity proteomics method for plasma proteomics. pubs.acs.org/doi/10.1021/...

Global cellular proteo-lipidomic profiling of diverse lysosomal storage disease mutants using nMOST A nanoflow multiomic workflow reveals how lysosomal dysfunction disrupts autophagy and mitochondrial homeostasis in LSD mutants.

Congrats to Yuchen He and Felix Kraus! This paper showcases fast multi-omics sample prep and our multi-omic single shot technology: Global cellular proteo-lipidomic profiling of diverse lysosomal storage disease mutants using nMOST | Science Advances #multiomics www.science.org/doi/10.1126/...

With the commercial BioIVT plasma we did see lower ID numbers than expected with Mag-Net, which could be due to more processing (fewer intact EVs). With patient samples in Fig. 7, Mag-Net acheives similar depth as the other Mag-Net Astral results Hamish shared below.

Hi! Yes as someone with a MS background it took me a while to understand how NPX works, it is basically sequencing counts put through multiple correction and normalization factors. I would recommend the Olink "Data normalization and standardization" white paper they have on their website.

I think the fact that Mag-Net targets EVs means that the more processed commercial plasma we used might not perform as well. For the patient samples in the small cancer study it gets much better depth!

Plasma has always been a difficult sample type but there are lots of really cool innovations that address this in different ways (including non-MS methods). We came up with several experiments to directly compare six of these methods and see where there might be differences.

A Technical Evaluation of Plasma Proteomics Technologies Plasma proteomic technologies are rapidly evolving and of critical importance to the field of biomedical research. Here we report a technical evaluation of six notable plasma proteomic technologies - ...

I'm excited to share that the first work of my PhD is now available on BioRxiv: A Technical Evaluation of Plasma Proteomics Technologies.

www.biorxiv.org/content/10.1...

The One Hour Human Proteome In BriefWe present deep analysis of the human proteome in less than 1 h. The Orbitrap Astral mass spectrometer, alongside advances in sample preparation, method development, and data processing, is ushering in a new era of proteomics where a harmony of throughput, depth, and sensitivity are enable studies of unprecedented impact. We contextualize these results within the landscape of human proteomics. Comparisons with previous instrumentation in interrogating gene knockouts demonstrate the potential of proteomics to elevate biological investigation in years to come.

New publication by Serrano et al. describes deep analysis of the human proteome in less than 1 hour! This paper describes the process and how it takes advantage of the new Orbitrap Astral mass spec high acquisition speeds. www.mcponline.org/article/S153...

For Maths fans, 2025 is a square.
45² = 45 x 45 = 2025
Also,
9² x 5² = 2025
40² + 20² + 5² = 2025
My favourite?
1³+2³+3³+4³+5³+6³+7³+8³+9³ = 2025
#Mathematics #teaching #education

🎉 Huge congratulations to the four incredible winners of the 2025 US HUPO Association Awards! 🏆 We're so excited to celebrate your achievements at US HUPO 2025 in Philadelphia! #USHUPO2025 #USHUPO

Has anyone used PCA as a way to compensate for sample effects in proteomics data? For example if large groups of proteins are changing in expression across different sample groups but probably not due to biology. Can you incorporate that principal component into a linear model to remove its effect?

I'm willing to bet almost every research paper from at least the past 6 months with 5+ authors has been aided by AI. Does something like using Chat-GPT to aid in R scripting necessitate an acknowledgement? What about using NotebookLM for bullet points of an introduction? What is acceptable right now