Using single-cell proteomics, we discovered protein (and functional) symmetry breaking in early mammalian embryos influencing cell fates. These proteomic asymmetries are detectable at the zygote stage, intensify by the 2-cell and 4-cell stage, and correlate with the sperm entry site, pointing to fertilization as a symmetry-breaking event. The clear protein differences allowed us to define two types of blastomeres, termed alpha and beta. ⬛ These differences predict the developmental potential of the blastomeres. Similar clustering and protein enrichment patterns found in human 2-cell embryos suggest this early asymmetry might be conserved. 1/

Single-cell proteomics is beginning to reveal functional coordination and principles. This motivates us to keep advancing the depth, accuracy, and scalability to further empower this analysis and help identify biological principles and targetable disease mechanisms.

Beyond heterogeneity, single-cell analysis can reveal functional coherence. Coherence, regularity and principles advance science. Full recording: piped.video/watch?v=z0dIhG4C…

Kicking off our first quarterly update from Parallel Squared Technology Institute (@ParallelSqTech): A boost in the throughput and quantitative accuracy of single-cell proteomics, and a warm welcome to our new Chief of Staff. Mass spectrometry based single-cell proteomics has advanced from "nearly impossible" to well demonstrated set of methods measuring thousands of proteins per cell, as traced in a recent Nature Technology Feature. A next big challenge: getting to higher throughput without losing depth or quantitative accuracy. 1/

Great to see multiplexed single-cell proteomics uncover new biology. We are looking forward to further increasing the plex and thus the power of the technology for biological discoveries! Congratulations to Luke Khoury and team !

Millions of PBMCs have been analyzed by scRNA-seq. Have we seen it all ? We profiled thousands of single PBMCs — proteomes and transcriptomes side by side — and uncovered functional coordination undetectable in the deepest scRNA-seq we could perform. biorxiv.org/content/10.64898… 1/

Among the numerous metrics evaluating academics, one stands out for me: Our intellectual heritage reflected in our students and colleagues, and their intellectual growth. That amplifies the initial sparks and bears more fruit than any one of us could have produced.

Excited to see what @slavov_n and @ParallelSqTech have been working on in @Nature's new article on single-cell proteomics.

The piece revisits the early days, our lab's initial work alongside other groups pioneering complementary approaches, together showing that mass-spec-based single-cell proteomics was feasible & promising. The proof-of-concept set the field in motion. 2/ nature.com/articles/d41586-0…

Nature just published a tech feature on single-cell proteomics. It highlights our journey from skepticism to robust technology and biological discoveries. I love that the feature starts with a biological discovery enabled by the technology: cell.com/cell/fulltext/S0092… 1/

High-throughput omics are powerful, but they don’t substitute for thinking. Discovering new biology often starts with intellectual input: frame sharp questions first, then choose the methods that can answer them, not the one that’s most convenient or currently fashionable. 1/

My personal experience is different. I always felt motivated to take risks, and my most recognized success is based upon taking risks. I recently talked about this experience: piped.video/z0dIhG4C3qI?si=9JDF…

What sets protein gradients in the liver ? The latest single-cell analysis from @ParallelSqTech shows it's often spatially regulated rates of protein synthesis and degradation. Liver cells (hepatocytes) form a gradient from the portal to the central vein, with variable protein abundance along this axis. Based on in vivo metabolic pulses with isotope labeled amino acids, we found zonated proteins whose mRNA levels don't track their protein levels: Their degradation rates do. A clear example is Mttp, a lipid-transfer protein that rises from portal to central cells. Its RNA doesn't explain that gradient. Its degradation rate does: Mttp is broken down faster in portal cells and more slowly in central cells. 🔷 The spatial pattern is set by protein stability. Zooming out, this wasn't a one-off. Across functional categories, zonated protein half-lives lined up with zonated protein abundance — a systematic signature of degradation shaping spatial identity. Interestingly, portal cells showed globally longer protein half-lives than central cells, especially for long-lived proteins. ⬛ A takeaway: Spatial identity in tissues isn't just about which genes get turned on — it's also about spatially regulated rates of protein degradation. Post-translational regulation is a driver of tissue architecture. 📄 Article: biorxiv.org/content/10.1101/…

These numbers are key to understanding RNA and protein analysis. The different counting statistics fundamentally shape technological challenges and opportunities. nature.com/articles/s41592-0… 1/

The much higher abundance of proteins brings dynamic range challenges but also supports much higher quantitative accuracy of single-cell MS proteomics, which enables different types of analysis. 2/ annualreviews.org/content/jo…

We have long aspired to realize the tech and bio opportunities of direct, accurate & scalable protein analysis: mcponline.org/article/S1535-… @ParallelSqTech is realizing these aspirations. Thanks to @Convergent_FROs, funders and colleagues for supporting our vision! 🚀

These results bode well for the feasibility of massively scaling up single-cell proteomics while preserving and increasing the depth of coverage and quantitative accuracy. 1/

That is why a team at @ParallelSqTech designed JMod. Not just for performance, but also for flexibility: 🔷 An open-source platform that the community can use to invent new acquisition strategies, experiment with creative designs, and develop the next generation of quantitative proteomics. 📄 JMod: Joint modeling of mass spectra for empowering multiplexed DIA proteomics 🔗 parallelsq.org/JMod

Accurately quantifying over 700K precursors in a single 9-plexDIA set of 20ng proteomes is a new milestone. JMod achieves high quantitative accuracy over a 32-fold dynamic range of proteome spiked-in ratios across single-cell and bulk sample sizes. 1/ doi.org/10.1101/2025.05.22.6…

The progress over next 5 years will be even faster 🚀 We are focussed on building the proteomics tools that will advance biological understanding and reveal the causes of disease.