iiinova — where your digital self keeps evolving

Johnny Yu @thejohnnyyu.bsky.social · 01/03/2025

Hey would a presentation on tahoe100 be a good one for this?

100

Johnny Yu @thejohnnyyu.bsky.social · 01/03/2025

We had a lot of fun with this project. From idea to preprint it was only 4 months! ⚡

000

Johnny Yu @thejohnnyyu.bsky.social · 01/03/2025

Team work makes the dream work!

000

Johnny Yu @thejohnnyyu.bsky.social · 01/03/2025

Just the start of a movement

000

Johnny Yu @thejohnnyyu.bsky.social · 01/03/2025

Yup split pool was an early key choice we made. Data is really not bad though, 1tb and you'll probably see some ultra dask type data structures to remove ram limitations super soon. So just solid state hdd and that'll be easy for a 1tb dataset

010

Johnny Yu @thejohnnyyu.bsky.social · 01/03/2025

We're excited! Stay tuned as this year several models built on this data are going to come out in quick succession 🙂

110

Johnny Yu @thejohnnyyu.bsky.social · 01/03/2025

Keep an eye on this - we're just getting started! Hope to have the ML community engaged as we continue in this direction 🚀

100

Reposted by Johnny Yu

Hani Goodarzi @genophoria.bsky.social · 25/02/2025

This was all made possible by the Mosaic platform! What is Mosaic? @thejohnnyyu.bsky.social took his work in our lab, and scaled it in every dimension… Mosaic brings a highly diversified, exquisitely optimized, and optimally balanced “cell village” approach to perturbation data collection.

121

Reposted by Johnny Yu

Hani Goodarzi @genophoria.bsky.social · 17/12/2024

Our latest from the indefatigable @thejohnnyyu.bsky.social in collaboration with Weissman and Shokat labs. Meet GENEVA, which enables simultaneous phenotyping and profiling of cancer cell drug responses at scale; both in vitro and in vivo across a variety of models: www.biorxiv.org/content/10.1...

biorxiv.org

Multiplexed mosaic tumor models reveal natural phenotypic variations in drug response within and between populations

Many agents that show promise in preclinical cancer models lack efficacy in patients due to patient heterogeneity that is not captured in traditional assays. To address this problem, we have developed...

13311

Johnny Yu @thejohnnyyu.bsky.social · 11/12/2024

This will be instrumental for data sets like our Tahoe 100 million. Especially as we scale into normalizing 100 million cell data sets

100

Reposted by Johnny Yu

Fabian Theis @fabiantheis.bsky.social · 11/12/2024

scRNA-seq data sets exploding in number and size - check out scanpy & anndata for >1b cells: new experimental update includes APIs for scaling with dask from anndata, integrated with lots of scanpy and rapids-singlecell functions. gist.github.com/ilan-gold/98...

25013