A full-stack institute for AI and biology research.
Palo Alto, CA
- Tweets 838
- Following 74
- Followers 47,669
- Likes 267
Virtual cell models are only as good as the data they learn from. To predict how any cell type responds to a perturbation, models need high-quality causal data generated across contexts, so technical noise doesn't masquerade as biology.
1
13
73
6,507
Here's what that looks like in practice. For this year's Virtual Cell Challenge, we ran identical Perturb-seq protocols in six cell lines, profiled 30M+ cells, then selected perturbations with ≥80% median knockdown.
Go behind the scenes in our blog :
arcinstitute.org/news/behind…
10
1,165
Congrats to the team on this release, and excited to announce our collaboration with @GoogleDeepMind -- more fun results to come!
3
15
172
23,464
In prophages, Unknown Group 27 reverse transcriptase loci hold arrays of roughly 150nt mcRNA units, diverse in sequence but sharing a predicted structure. Expressed in E. coli, all five systems reverse transcribed the central hairpin of each unit into 50-100nt ssDNA.
1
1
4
803
Minerva is built for researchers looking for non-coding elements and RNA-protein systems that homology searches can’t reach. Code and Minerva-MLM checkpoints and Colab notebooks are available for predictions and fine-tuning.
Now on @biorxivpreprint: biorxiv.org/content/10.64898…
6
767
In Pseudomonas, the TwoAYGGAY RNA family extends well past its Rfam annotation, with three further hairpins and a 3’ pseudoknot on an elongated basal stem. A covariance model rebuilt on the Minerva predictions found 16,086 matches across 1,148 of 1,324 strains.
1
3
329
For any portion of a genome, Minerva returns a map of which positions pair with which, including pseudoknots and overlapping hairpins that are hard to capture otherwise. Nothing needs to be aligned first, and one pass per region is fast enough to scan a whole genome.
1
8
1,377
Many important bacterial non-coding RNAs and structural elements remain undiscovered. To help, @_David_Li, @garykbrixi, @mfgrp, @BrianHie & team introduce Minerva, which uses a genome language model to predict RNA base pairing, repeats, & other interactions from sequence alone.
5
26
175
19,087
As with biological design, AI can also accelerate biological discovery.
In new work led by @_David_Li and @garykbrixi with @mfgrp, we show how genome language models enable systematic discovery of genetic elements, revealing a new class of reverse-transcriptase mechanisms.
4
61
343
32,126
Arc Institute retweeted
New issue of @CellCellPress out today on AI and biology, featuring several papers from @arcinstitute: State cell perturbation model, scBaseCount AI agent curated single-cell dataset, Virtual Cell Challenge 2026 commentary, and Tahoe-100M from our friends at Tahoe Bio and featured in Arc Virtual Cell Atlas. Congrats to the teams. Lots more exciting AI/bio work cooking at Arc Institute! cell.com/cell/current
1
8
21
6,075
Arc Institute retweeted
The data behind scBaseCount were already public. The challenge was making them findable, interpretable, and comparable
Our new @CellCellPress paper describes how an AI agent helped build a uniformly processed resource of >502M cells
doi.org/10.1016/j.cell.2026.… @arcinstitute
1
10
38
4,249
Want to join Arc? We're hiring experts in disease bio, computational bio, functional genomics, and NGS for our Tech Centers, plus postdocs and research associates in our Core Labs. All roles are based in Palo Alto, and are on-site or hybrid.
Learn more or apply here: arcinstitute.org/jobs
1
18
96
14,897
Arc Institute retweeted
I am honored to have been selected as one of MIT Technology Review's Innovators Under 35 this year.
🔗ter.li/05p3ttfg
🔗technologyreview.com/superto…
This recognition is shared with the many amazing people I get to work with everyday as a @bioe_stanford PhD student and at @arcinstitute. Doing research alongside my brilliant labmates and advisor @BrianHie has been the best part of my PhD, and the science is truly a product of that.
My research has been driven by the insight that biology operates on design principles that can be leveraged to transform medicine, sustainability, materials, and beyond, and that computational modeling, including AI, will help us decipher those design principles at scale. Nature's innovations are written in genomes, and we have a long way to go in understanding them. You can read more about some of our efforts developing AI to understand and design genomes in my comment below.
Thank you to everyone at the @techreview and all the judges for organizing this, especially @AmyNordrum and @antonioregalado, for your support. I can't wait to see all the incredible things the other Innovators do!
17
13
105
16,033
Loss of function for 1 gene can make a cancer cell resist a DNA damage response (DDR) inhibitor, while another gene can make it easy to kill. Core Investigator @LukeGilbertSF and collaborators mapped both across five DDR inhibitors and found loss of PRDX1 sensitizes cells to all five.
5
13
50
7,916
The team then repeated the screens in PRDX1 knockout cells. Knocking down IREB2, PAX7, or MRGBP restored growth under DNA-PK inhibition and lowered γH2AX. All three lowered labile iron, and chelating iron partly rescued the cells, indicating PRDX1 protects the genome from iron-driven oxidation.
1
1
667
PRDX1 clears hydrogen peroxide that free iron turns into DNA-damaging hydroxyl radicals. This work identifies PRDX1 and its upstream genes as potential druggable targets to pair with DDR inhibitors or to use in cancers already deficient in DNA repair.
See in @nchembio: nature.com/articles/s41589-0…
1
3
660