Rahul Satija

@satijalab · joined 23 Sep 2009

Core Member, New York Genome Center; Professor, Biology, NYU

28 199Followers

352Following

963Posts total

167.2KViews on collected posts

โพสต์ล่าสุด

Rahul Satija@satijalab25 Aug

Excited to share a public beta of Seurat 5.6, implementing major improvements for speed/memory/scalability of the core workflow, especially for large datasets. Try it out! Installation instructions at: https://t.co/4usY1DOAyW 20.2K views · 172 likes · 29 reposts · 1 replies Open on X →

Rahul Satija@satijalab23 Jul

Excited to share Pan-human Azimuth, an NIH @_hubmap reference for single-cell data https://t.co/PzQ1Z7Sccl We invested deeply in human curation of training data, prioritized diversity+breadth+extensive QC, and uniformly labeled cells across 23 tissues in a single hierarchy (1/) 20.5K views · 176 likes · 46 reposts · 5 replies Open on X →

Rahul Satija@satijalab22 Jun

@satijalab Kudos! Could you also please help the field understand that they shouldn't use FindMarkers for DEG analysis -- the time has come to move beyond pseudoreplication (n=# of cells) and associated false positives/inflated significance. A warning message would be helpful. 1K views · 12 likes · 1 reposts · 0 replies Open on X →

Rahul Satija@satijalab22 Jun

We're really excited to see how these types of improvements will improve the efficiency of single-cell and spatial analysis -especially for large datasets! 2.7K views · 15 likes · 0 reposts · 0 replies Open on X →

Rahul Satija@satijalab22 Jun

We're excited to work with them to see if our improvements can further synergize with theirs. We look forward to updating the official Seurat release soon, but in the mean time you can check out/try AutoZyme at: https://t.co/nESOPmQx1m 4.4K views · 24 likes · 3 reposts · 2 replies Open on X →

Rahul Satija@satijalab22 Jun

Over the past few months, we've been rewriting core workflows in Seurat with coding agents to improve speed and scalability. By moving steps from R to C++ and optimizing code, we've seen huge improvements - in some cases 10-fold and above - with identical outputs. (1/) 49.1K views · 293 likes · 31 reposts · 6 replies Open on X →

Rahul Satija@satijalab22 Jun

We are preparing a public release - and were delighted to see fantastic new work from @c_kendziorski and @Elliotxie_ . Their findings mirror our internal results - but go further in establishing an awesome agentic workflow to optimize scientific code: https://t.co/EWbdMxNLp1 5.5K views · 62 likes · 6 reposts · 1 replies Open on X →

Rahul Satija@satijalab12 Jun

Thanks to @muronglizi for the last talk of #singlecellgenomicsday, demonstrating how Multimodal AI models can help explore and analyze massive single-cell and spatial data using natural language. And thats a wrap for Single Cell Genomics Day 2026- see you next year! 3.3K views · 12 likes · 2 reposts · 0 replies Open on X →

Rahul Satija@satijalab18 Feb

@satijalab 👏👏👏 201 views · 1 likes · 0 reposts · 0 replies Open on X →

Rahul Satija@satijalab17 Feb

Congratulations and many thanks to lead authors @alex_bradu (who runs our single cell genomics day!), and John Blair (@jblairsci.bsky.social). And to our funders @genome_gov (via @cegs_ica), @SMaHTnetwork, and @cziscience 1.6K views · 13 likes · 0 reposts · 0 replies Open on X →

Rahul Satija@satijalab17 Feb

Learn more and try it out at https://t.co/pZhYQb0jAv And lots more to come! With @CalebLareau’s lab, we’ve shown that VIPerturb-seq is compatible with RNA-based enrichment (#VIPerff), and we’re excited to share this soon. 1.8K views · 13 likes · 0 reposts · 1 replies Open on X →

Rahul Satija@satijalab17 Feb

We hope that VIPerturb-seq will be useful for data platforms and virtual cell builders. But we really hope it will help individual labs with specific research questions, even if they don’t have huge sequencing budgets. 1.3K views · 10 likes · 0 reposts · 1 replies Open on X →

Rahul Satija@satijalab17 Feb

Of course, VIPerturb-seq provides scRNA-seq data for each cell, so we can molecularly characterize new regulators that we identify. We learn a ‘Ragulator’-perturbation module, and use this to assess the molecular function of additional hits. https://t.co/K1BGdF9TJQ

1.3K views · 8 likes · 1 reposts · 1 replies Open on X →

Rahul Satija@satijalab17 Feb

When we profile the VIPs, we recover essentially the entire Rag–Ragulator-FLCN complex, which senses amino acid availability. VIPs are enriched at ~10-60 fold. This links nutrient-sensing to VIM regulation, which we validated with independent intracellular measurements as well.

1.4K views · 8 likes · 0 reposts · 1 replies Open on X →

Rahul Satija@satijalab17 Feb

One more challenge: Even with cheaper library prep, sequencing costs are still massive, especially for 1M+ cell screens So we do a second fun thing! Since VIPerturb-seq is compatible w/fixation, we can phenotypically enrich for ‘Very Important Perturbations’ prior to sequencing

2.3K views · 12 likes · 3 reposts · 1 replies Open on X →

Rahul Satija@satijalab17 Feb

We repeated a genome-wide GuEST-List infection, but fixed/stained and sorted (top 3%) on intracellular vimentin levels, so only negative regulators are VIPs. This time we only profiled 12,000 cells. So its a genome-wide Perturb-seq screen - but we sequenced it on a NextSeq! 1.4K views · 8 likes · 0 reposts · 1 replies Open on X →

Rahul Satija@satijalab17 Feb

Inspired by @JswLab, we generated a mini Genome-wide Perturb-seq, using just two 10x lanes (!). Far too much data for one tweet (or one Figure), but it works beautifully. The ability to assess the molecular function of every gene in an afternoon is mind-boggling https://t.co/yh

12.2K views · 181 likes · 32 reposts · 1 replies Open on X →

Rahul Satija@satijalab17 Feb

Using GuEST-List with probe-based scRNA-seq lets us do some fun things! First, we take advantage of combinatorial barcoding (Flex v2) where multiple pre-indexed cells can load into one droplet. We capture 440,000 cells in a single reaction lane, a massive increase in throughput

1.6K views · 8 likes · 0 reposts · 1 replies Open on X →

Rahul Satija@satijalab17 Feb

This yields the highest quality Perturb-seq data my lab has generated. Compared to conventional Perturb-seq with RT-based scRNA-seq, we see an increase of ~65% in genes/cell We also detect >100 gRNA UMI/cell, and make assignments in 97% of cases https://t.co/TlFTnlBmUl

2K views · 17 likes · 2 reposts · 1 replies Open on X →

Rahul Satija@satijalab17 Feb

We use this to design a barcoded ‘Guides for Enrichment-based Screening of Targets Library’, or GuEST-List GuEST-List is based on Dolcetto and includes CRISPRi gRNA for all genes, even if only a few later emerge as being ‘Very Important Perturbations’ :) https://t.co/z8R48fyksa

2.8K views · 11 likes · 0 reposts · 1 replies Open on X →

Rahul Satija@satijalab17 Feb

VIPerturb-seq is based on a simple idea: we associate CRISPR gRNA with a split barcode detected by the combination of a left-hand + right-hand probe This lets us easily read out very large (genome-wide) gRNA libraries w/probe-based scRNA-seq, like the Flex kit from @10xGenomics

4.2K views · 22 likes · 2 reposts · 1 replies Open on X →

Rahul Satija@satijalab17 Feb

Excited to share VIPerturb-seq! New tech from my lab which aims to improve the cost, data quality, and efficiency of single-cell CRISPR screens so that they are accessible to any lab - even at genome-wide scale Preprint and 🧵 (1/): https://t.co/m8nleniSUD 26.4K views · 322 likes · 84 reposts · 5 replies Open on X →

เทียบกับบัญชีขนาดเดียวกัน

2 โพสต์จาก 90 วันที่ผ่านมา เทียบกับช่วง 10K–100K ผู้ติดตาม แสดงในวงกว้าง แต่มีผู้ชมตอบสนองน้อย.

ยอดดูมัธยฐาน20 380บัญชีนี้924ค่ามัธยฐานของ 10K–100K

การเข้าถึง, %72.27%บัญชีนี้3.62%ค่ามัธยฐานของ 10K–100K

การมีส่วนร่วม, %1.06%บัญชีนี้1.52%ค่ามัธยฐานของ 10K–100K

ตัวชี้วัด	บัญชีนี้	ค่ามัธยฐานของ 10K–100K	อัตราส่วน
ยอดดูมัธยฐานต่อโพสต์	20 380	924	22.1×
การเข้าถึง (ยอดดู ÷ ผู้ติดตาม)	72.27%	3.62%	20.0×
อัตราการมีส่วนร่วม	1.06%	1.52%	0.70×

บัญชีอื่นในช่วงนี้ → เปรียบเทียบกับบัญชีอื่น → ค่าอ้างอิงเหล่านี้คำนวณอย่างไร →

Growth & engagement

How the posts we collected actually performed: views and reaction rate post by post, what the audience did with them, and where the follower count goes.

Views per post

1.4K17 Feb

1.3K

1.8K

1.6K

20118 Feb

3.3K12 Jun

5.5K22 Jun

49.1K

4.4K

2.7K

20.5K23 Jul

20.2K25 Aug

Last 14 collected posts, oldest on the left. The scale is logarithmic: one post can outrun the rest a hundred times over.

Engagement rate per post

0.67%17 Feb

0.78%

0.87%

0.77%

0.81%

0.50%18 Feb

0.43%12 Jun

1.29%22 Jun

0.68%

0.66%

0.55%

1.24%

1.12%23 Jul

1.01%25 Aug

Reactions — likes, reposts, replies and quotes — divided by views. Median for 10K–100K accounts is 1.52%.

What the audience does

Likes62.3%1 400 in total

Reposts10.8%242 in total

Replies1.4%31 in total

Quotes0.6%13 in total

Bookmarks25.0%561 in total

Share of every reaction we collected for this account. Replies mean argument, reposts mean endorsement, bookmarks mean the post was worth keeping.

The follower curve appears once this account has two daily snapshots — we take one a day, and this one is on its first.

โพสต์ล่าสุด

เทียบกับบัญชีขนาดเดียวกัน

Growth & engagement

Views per post

Engagement rate per post

What the audience does

บัญชีที่คล้ายกัน