Nikolai Slavov
@slavov-n.bsky.social
2.4K followers 130 following 700 posts
Mentor, scientist & engineer. Having fun in @slavovlab.bsky.social and Parallel Squared Technology Institute @parallelsq.bsky.social with biology & single-cell proteomics. https://nikolai.slavovlab.net
Posts Media Videos Starter Packs
Pinned
slavov-n.bsky.social
mRNA-mRNA correlations across the single cells from a tissue are rarely interpreted.

They differ from the corresponding protein-protein correlations?

𝐖𝐡𝐲 ❓

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

🧵
slavov-n.bsky.social
A new article in nature shows linear increase accumulation in sperms cells with aging.

On average, 1.67 mutations are added per year per haploid genome.
slavov-n.bsky.social
Academia is not about publishing papers.

It’s about creating knowledge and teaching it to the world.

It’s about asking big bold questions and mentioning the next generation of intellectual leaders.

The difference makes a huge difference.
slavov-n.bsky.social
Perturb-seq datasets show a clear difference in emphasis:

1️⃣ Tahoe aimed to maximize cell numbers

2️⃣ X-Atlas/Orion aimed to maximize coverage depth

What are the tradeoffs of these strategies ?

What inferences benefit from 1️⃣ vs. 2️⃣ ❓
slavov-n.bsky.social
The counting statistics for single-cell MS support accurate estimates of protein abundance in single cells, which is essential for quantitative analysis.

Further, measuring and modeling protein degradation is key to biological interpretations.

pubs.acs.org/doi/10.1021/...
Transformative Opportunities for Single-Cell Proteomics
Many pressing medical challenges, such as diagnosing disease, enhancing directed stem-cell differentiation, and classifying cancers, have long been hindered by limitations in our ability to quantify p...
pubs.acs.org
slavov-n.bsky.social
What was once a projection is now reality.

Exciting times for quantitative single-cell analysis !
Reposted by Nikolai Slavov
parallelsq.bsky.social
It's good to look back and see how far we have come since Research Fest took place almost six months ago.

Over the next few weeks, we will post some throwback videos from the presentations. Let's start with opening remarks by PTI's Director, @slavov-n.bsky.social:

youtu.be/Sc2QLXb84XI
Proteomics: The arc of progress | Nikolai Slavov
YouTube video by Parallel Squared Technology Institute
youtu.be
slavov-n.bsky.social
Earlier this year, @parallelsq.bsky.social hosted a research fest.

We began with a broad perspective:
◾️ A century of remarkable progress! Proteomics drove conceptual discoveries and medical treatments.

Listen to what's next!

Proteomics: The arc of progress
youtu.be/Sc2QLXb84XI?...
slavov-n.bsky.social
AlphFold is an example of supervised learning.

We first train on subsets of problems for which the answer is known, and then apply the models to other subsets to generate hypotheses, hopefully reliable and useful ones.
slavov-n.bsky.social
In this article, I shared my perspective on AI.

AI excels at supervised learning, and well annotated datasets can help unlock new AI capabilities.

🔗 www.nature.com/articles/s42...

OA 🔗 rdcu.be/eI4JW
slavov-n.bsky.social
𝐏𝐫𝐨𝐭𝐞𝐚𝐬𝐨𝐦𝐞𝐬 𝐩𝐥𝐚𝐲 𝐤𝐞𝐲 𝐫𝐨𝐥𝐞𝐬 𝐢𝐧 𝐧𝐞𝐮𝐫𝐨𝐝𝐞𝐠𝐞𝐧𝐞𝐫𝐚𝐭𝐢𝐨𝐧

Neuronal inflammation induces PSMB8

Integrating PSMB8 into neuronal proteasomes reduces their activity

This causes PFKFB3 accumulation, a metabolic switch, and ferroptosis

🧵
slavov-n.bsky.social
Recently, Sam Altman stated that limited compute may force OpenAI to choose between curing cancer and educating the world.

Now, they are prioritizing their limited compute towards another TikTok-like app.

Tough choices.
slavov-n.bsky.social
I am looking forward to ICSB next week.

I will present:
1⃣ New technology from @parallelsq.bsky.social:
parallelsq.org/blog/proteom...

2⃣ New biology from @slavovlab.bsky.social
biorxiv.org/content/10.1...
slavov-n.bsky.social
The result from the above example generalizes across all proteins that we measured.

🔗 www.biorxiv.org/content/10.1...
slavov-n.bsky.social
Let's first zoom in on an example:

Protein-protein correlation: 0.56
mRNA-mRNA correlation: 0.01

⬛️ The proteins are strongly regulated by degradation.
slavov-n.bsky.social
mRNA-mRNA correlations across the single cells from a tissue are rarely interpreted.

They differ from the corresponding protein-protein correlations?

𝐖𝐡𝐲 ❓

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

🧵
slavov-n.bsky.social
Interesting questions:

Why long-lived proteins exist in metabolically active cellular environments ?

How are these proteins maintained over time ?
slavov-n.bsky.social
The number depends on the desired precision.

The counting error is 1 / sqroot(n), where n is the number of counted molecules.

If we want 10 % counting error, we need to go through 100 bottles, so 1 / sqroot( 100 ) = 0.1.
slavov-n.bsky.social
These 3-L bottles contain one million tiny colored spheres each.

One sphere is black (1 ppm).

Finding the black sphere is comparable to detecting a protein present at ~ 6,000 copies in the proteome of a human cell.

Quantifying the protein requires analyzing multiple jars.
slavov-n.bsky.social
An amazing feat of regulation.

Two very different regulatory mechanisms converge to shape the mitochondrial proteome and keep us alive!
slavov-n.bsky.social
I have always considered creativity as a core skill in scientific research. Part of the challenge is perhaps in the variability of how different people define ideas and novelty.

Thanks for raising awareness of it and helping colleagues sharpen their skills.

blog.slavovlab.net/2015/05/31/i...
Ideas
I have heard many biomedical researchers express the opinion: “In biology, ideas are cheap. It is the doing that matters.” I do not share this opinion. However, I would like to understa…
blog.slavovlab.net