1/12 Excited to share our team's latest work and the first @xaira_thera preprint! Here, we introduce FiCS Perturb-seq, an industrialized platform for generating scaleable, high-quality perturbation data. 📄 Read the preprint:

Virtual Cell community - this one's for you! X-Atlas/Orion is now live on Hugging Face. Train your own models with streamlined workflows built into the Hugging Face API. 🔗 HuggingFace: https://t.co/U2BGIBGx4w 📜 License: cc-by-nc-sa-4.0

New to virtual cells? @freethink’s primer explains these AI-powered, in silico models—why they’re game-changers for biology & the hurdles still to solve. Worth a read

In the wake of all the Bio x AI updates, I'm also here to spread awareness of important cat-related research. Isparta et al analyzed 400+ cat videos and found that cats favor the leftward side (p < 0.001), allowing for fast response to stimuli. Paper: https://t.co/uahuvZcHBG

We are looking for talented postdocs/visiting students to join our team @MITdeptofBE and @kochinstitute to innovate spatial multi-omics technologies, map clinical samples, and integrate AI/ML models to advance next-generation cancer immunotherapy. Please help to spread!!

Such an exciting initiative. Congrats to all those involved! Hope people put the X-Atlas/Orion dataset to use 💪

12/12 This work is the result of an incredible collaboration between @Xaira_Thera and @ForesiteLabs! Many thanks to all the amazing co-authors (see preprint for full list): @Stanley_TH @jiangzhuzime @qtaznangel @awblocker and @inCiChu

11/12 In conclusion, FiCS Perturb-seq is a robust platform for executing large-scale perturbation screens, enabling the creation of datasets like X-Atlas/Orion to accelerate causal foundation models in predictive biology. 📊 Download X-Atlas/Orion:

10/12 As part of this work, we’re releasing X-Atlas/Orion, the largest publicly available Perturb-seq atlas to date! It contains ~8 million cells deeply sequenced to >16k UMIs per cell from the two genome-wide FiCS Perturb-seq screens described above.

9/12 This approach moves beyond treating genetic perturbations as simple on vs off switches, enabling a more nuanced understanding of how gene dosage impacts cellular responses. Indeed, we observed that more sgRNA leads to stronger cellular responses in the same perturbations.

8/12 Excitingly, we find that sgRNA abundance can be used as a reliable proxy for gene knockdown efficiency. This enables the dissection of dose-dependent genetic effects with unprecedented precision.

7/12 Data generated using FiCS Perturb-seq is biologically meaningful. We show it validates physical protein-protein interactions and clusters perturbations into known biological complexes, such as ribosome biogenesis, protein synthesis, and Mediator complex.

6/12 Furthermore, FiCS Perturb-seq also delivers remarkably consistent data with significantly lower batch-to-batch variation compared to previous atlases, ensuring consistency and reliability across experiments. This means better data for more robust models!

5/12 A major bottleneck of perturbation data generation is the need to use fresh cells. Using FiCS Perturb-seq, we show that cryopreservation of fixed cells maintained RNA-seq quality for up to 140 days, decoupling cell dissociation from library preparation.

4/12 Using FiCS Perturb-seq, we performed two genome-wide screens in HCT116 and HEK293T cell lines. Both screens had better sensitivity compared to existing genetic and chemical perturbation atlases, detecting a greater number of genes and UMIs per cell.

3/12 To overcome these challenges, we developed FiCS Perturb-seq, a platform that integrates fixation, FACS enrichment, cryopreservation, superloading, and automation for scalable, high-quality Perturb-seq data generation.

2/12 This work was inspired by the need for large, high-quality perturbation datasets to train foundation models that understand causal relationships of biology. Current methods for generating perturbation data are limited by throughput and batch-to-batch variability.

Today, we're releasing the fuel for the next generation of AI in biology 🧬 X-Atlas/Orion is now the largest public genome-wide Perturb-seq dataset, built to create better “virtual cell” models and accelerate drug discovery. Learn more:

Excited to share that we are emerging from stealth mode today! Stay tuned for updates as we use AI to transform the drug discovery process

I am often asked by scientists across fields for advice & resources on how to get started with #scRNAseq analyses I summarize here such an analysis into a flowchart with 9 core steps. Analyzing your own scRNAseq data today is more accessible than ever And it gets even easier🧵