Pinned Tweet
We’re recruiting!
The Bo Xia Lab @broadinstitute @MassGenBrigham @harvardmed is looking for postdocs, students, and visitors to build AI-powered predictive & programmable biology.
If you’re excited about multimodal AI × genome regulation × cell fate, join us!
🔗 httpringNow #PhD #postdocs #AIinBiology #Genomics #GeneRegulation #CellFate
The Bo Xia Lab @broadinstitute @MassGenBrigham @harvardmed is looking for postdocs, students, and visitors to build AI-powered predictive & programmable biology.
If you’re excited about multimodal AI × genome regulation × cell fate, join us!
🔗 httpringNow #PhD #postdocs #AIinBiology #Genomics #GeneRegulation #CellFate
6
149
728
62,221
🤯🤯🤯 what an amazing era to live in!! congratulations Levent!!
hello there the jacobian conjecture is false thanx to my close friend akhil for asking about it and my other close friend fable for working during the world cup final
((1+xy)^3 z + y^2 (1+xy) (4+3xy), y + 3 x (1+xy)^2 z + 3 x y^2 (4+3xy), 2 x - 3 x^2 y - x^3 z): \C^3\to \C^3, has jacobian determinant -2, and sends (0, 0, -1/4), (1, -3/2, 13/2), and (-1, 3/2, 13/2) to (-1/4, 0, 0)
0
0
6
1,898
Congratulations @jengreitz and the team on the formal publication of this important milestone of regulatory genomics!
An encyclopedia of enhancer-gene regulatory interactions — online today! nature.com/articles/s4158…
Now with an improved model, expanded maps across 1400+ biosamples, larger validation CRISPR datasets, and guidance on applying the model
1/
0
4
35
8,532
Bo Xia retweeted
How much of the human genome is essential?
Two pieces out today from our lab: 1) a method to map essential genomic intervals at gigabase scale, and 2) an argument that it's time to consider synthesizing a minimal human genome.
biorxiv.org/content/10.648…
nature.com/articles/d4158…
Two pieces out today from our lab: 1) a method to map essential genomic intervals at gigabase scale, and 2) an argument that it's time to consider synthesizing a minimal human genome.
biorxiv.org/content/10.648…
nature.com/articles/d4158…
8
49
180
43,262
Congratulations @landau_lab and the teams! It was amazing to see this exciting work coming out mapping TF bindings in single cells. Glad to see that our C.Origami with @tan_jimin and @artsinyc can contribute to investigate the cis-regulatory mechanisms mediated by proteins.
Genome regulation operates on (mostly) identical sequences across cell types/states, with chromatin proteins executing such cell type specific functions. So much work can be enabled by the single-cell approaches like D&D-seq that map TF binding along the genome! Congratulations again to the team!
Genome regulation operates on (mostly) identical sequences across cell types/states, with chromatin proteins executing such cell type specific functions. So much work can be enabled by the single-cell approaches like D&D-seq that map TF binding along the genome! Congratulations again to the team!
Exciting breakthrough technology from the lab, now live in @CellCellPress ! Instead of cutting the genome where proteins bind (e.g., Cut&Tag), D&D-seq scars the DNA with a deaminase, allowing single cell genome mapping of TFs and chromatin remodellers!
2
8
62
11,041
Bo Xia retweeted
Going from the discovery of epigenetic silencing of CTCF insulators nature.com/articles/s4158… to aberrant E-P looping cell.com/molecular-cell… to a successful clinical trial nature.com/articles/s4159… all from Brad Bernstein's lab
0
28
127
11,351
Bo Xia retweeted
Today we're announcing ESMFold2, an open scientific engine to power prediction, design, and discovery across protein biology.
The new model delivers state of the art performance on protein interactions, especially antibodies, a critical modality for therapeutics.
We have miniprotein binders and single chain antibodies across five therapeutic targets that are important in cancer and immunology. We are seeing very high success rates, and affinities at levels consistent with therapeutic activity.
We’re also releasing an atlas of 6.8 billion proteins, and 1.1 billion predicted structures.
ESMFold2 is built on a state of the art language model that has been trained on billions of protein sequences.
A world model of protein biology emerges through language modeling.
We’ve used the techniques of mechanistic interpretability developed to understand large language models to understand the concepts ESM uses to represent proteins.
The model’s representation space has a compositional organization of features across scales, levels of complexity, and abstraction, that reflects and mirrors the understanding of protein biology developed through a century of empirical science.
This understanding emerges without prior knowledge, just from language modeling of protein sequences.
Language models are becoming a powerful substrate to understand and program biology.
The design of protein interactions is one of the most fundamental problems in biophysics, and has critical implications for the discovery of new medicines. A simple gradient based search with the model was able to discover high-affinity protein binders.
I'm excited by the potential this has to accelerate basic science and the understanding of proteins. And especially for the new avenues it opens up for therapeutic design and medicine.
The new model delivers state of the art performance on protein interactions, especially antibodies, a critical modality for therapeutics.
We have miniprotein binders and single chain antibodies across five therapeutic targets that are important in cancer and immunology. We are seeing very high success rates, and affinities at levels consistent with therapeutic activity.
We’re also releasing an atlas of 6.8 billion proteins, and 1.1 billion predicted structures.
ESMFold2 is built on a state of the art language model that has been trained on billions of protein sequences.
A world model of protein biology emerges through language modeling.
We’ve used the techniques of mechanistic interpretability developed to understand large language models to understand the concepts ESM uses to represent proteins.
The model’s representation space has a compositional organization of features across scales, levels of complexity, and abstraction, that reflects and mirrors the understanding of protein biology developed through a century of empirical science.
This understanding emerges without prior knowledge, just from language modeling of protein sequences.
Language models are becoming a powerful substrate to understand and program biology.
The design of protein interactions is one of the most fundamental problems in biophysics, and has critical implications for the discovery of new medicines. A simple gradient based search with the model was able to discover high-affinity protein binders.
I'm excited by the potential this has to accelerate basic science and the understanding of proteins. And especially for the new avenues it opens up for therapeutic design and medicine.
78
445
1,644
634,796
Bo Xia retweeted
Could the folding of synthetic gene circuits in 3D shape how genes are expressed? Today @ScienceMagazine we report on the role of gene syntax in shaping feedback between transcriptional activity and genome folding for advanced circuit design🧵 (1/n)
23
187
689
73,371
Bo Xia retweeted
NO, THIS ARE NOT QUESTIONS THAT KEEP PHD UP AT NIGHT.
It's a good demonstration/eval that demonstrate LLMs can synthesize expertise from different domains and write code to efficiently solve problems, though.
Give a cell biologist an IGV and they will be perfectly fine at this
It's a good demonstration/eval that demonstrate LLMs can synthesize expertise from different domains and write code to efficiently solve problems, though.
Give a cell biologist an IGV and they will be perfectly fine at this
Anthropic just dropped a benchmark that should make every scientist pay attention. BioMysteryBench puts AI models through 99 real bioinformatics challenges, using raw, messy datasets from actual research, think unprocessed DNA sequences and clinical samples.
However: these s with neat answers. They're the kind of open-ended puzzles that keep PhD students up at night.
The results are exciting. Claude's latest models (4.7) solve the majority of tasks that trained human experts can handle, and on 23 problems that a panel of five domain experts couldn't crack, Claude Mythos Preview nailed 30% of them. How? By combining knowledge from hundreds of thousands of papers and layering multiple analytical strategies when uncertain, essentially doing what a room full of specialists would do, but faster and in a single run.
Genentech and Roche independently confirmed this trajectory with their own CompBioBench, where Claude Opus 4.6 reached 81% overall accuracy and 69% on the hardest questions.
Two separate benchmarks, same conclusion: AI is no longer just keeping pace with biologists, it's pulling ahead on some of the hardest problems.
6
5
42
9,028
Bo Xia retweeted
How does a cell learn? In our new perspective in @Nature we propose a model where the properties of the AP-1 family of transcription factors – stress-induced feedback, regulatory combinatorics & cellular memory – encode a mechanism for cellular learning. nature.com/articles/s4158…
6
68
224
23,503
Bo Xia retweeted
Luminal surface proteome of the brain vasculature uncovers blood-brain barrier regulators @ScienceMagazine:
science.org/doi/10.1126/sc…
Big thanks to Zijian, Zuzhi, Yupu, Khanh, Yuxiang, and all collaborators to make this possible!
science.org/doi/10.1126/sc…
Big thanks to Zijian, Zuzhi, Yupu, Khanh, Yuxiang, and all collaborators to make this possible!
8
31
108
73,914
Bo Xia retweeted
Form follows function, or function dictates form?
📢Preprint: we build Perturb-CLEAR, in vivo screen of neuroanatomy! Disease perturbations produce selective morphology-RNA changes that RNA impact alone cannot predict. Was led by the invincible @BoliWu!
biorxiv.org/content/10.648…
📢Preprint: we build Perturb-CLEAR, in vivo screen of neuroanatomy! Disease perturbations produce selective morphology-RNA changes that RNA impact alone cannot predict. Was led by the invincible @BoliWu!
biorxiv.org/content/10.648…
7
50
185
21,817
Bo Xia retweeted
New preprint! @JShendure @CXchengxiangQIU Can we learn regulatory grammars of human cell types we'll never be able to profile - by training on mouse development and transferring across 241 mammalian genomes?biorxiv.org/content/10.648…
1
36
146
13,336
Bo Xia retweeted
It’s well known that inflammation increases cancer risk, but how?
The answer: the epigenome "remembers" inflammation and primes stem cells for cancer.
Here is our paper: nature.com/articles/s4158…
And a special shoutout to the lead author @snaga13
A 🧵
The answer: the epigenome "remembers" inflammation and primes stem cells for cancer.
Here is our paper: nature.com/articles/s4158…
And a special shoutout to the lead author @snaga13
A 🧵
15
129
563
42,993
Bo Xia retweeted
🚨🧬🩸 Delighted to have this @NatureRevGenet review w/ @michaelpoeschla published today - Genetic Influences on #hematopoiesis: rdcu.be/e9vlD
6
68
229
28,305
Bo Xia retweeted
Excited to share: enhanced capture in vivo multiome Perturb-seq!
Worried about low gRNA recovery and waste reads on cells that can’t assign a perturbation? We engineered gRNA transcripts to stay in the nucleus, enabling efficient multiomic Perturb-seq!
biorxiv.org/content/10.648…
Worried about low gRNA recovery and waste reads on cells that can’t assign a perturbation? We engineered gRNA transcripts to stay in the nucleus, enabling efficient multiomic Perturb-seq!
biorxiv.org/content/10.648…
1
39
213
14,512
Bo Xia retweeted
Thrilled to announce alphagenome-pytorch, an accurate, readable, and careful port of AlphaGenome's architecture and weights to PyTorch. Work with @gtcaa @m_kjellberg @chriswzou @TuXinming as part of the GenomicsxAI initiative between @anshulkundaje and @pkoo562 labs.
6
106
445
65,459
Bo Xia retweeted
Evo 2, our genome language model that generalizes:
- across biological prediction and design tasks,
- across all modalities of the central dogma,
- across molecular to genome scale, and
- across all domains of life,
is published today in @Nature.
- across biological prediction and design tasks,
- across all modalities of the central dogma,
- across molecular to genome scale, and
- across all domains of life,
is published today in @Nature.
10
74
378
60,903
Bo Xia retweeted
Masahiro Nagano and I were fortunate to collaborate with the Vos&Farnung labs:
CTCF dimerization drives higher order nucleosome assemblies.
Disrupting CTCF dimerization in cells weakens CTCF loops and causes growth + differentiation defects.
biorxiv.org/content/10.648…
CTCF dimerization drives higher order nucleosome assemblies.
Disrupting CTCF dimerization in cells weakens CTCF loops and causes growth + differentiation defects.
biorxiv.org/content/10.648…
1
23
89
6,732
Bo Xia retweeted
Too much data, too little thinking.
A important essay from Ruslan Medzhitov on the importance of understanding data, not just generating it. A must read.
@RMedzhitov @YaleIBIO
nature.com/articles/s4157…
A important essay from Ruslan Medzhitov on the importance of understanding data, not just generating it. A must read.
@RMedzhitov @YaleIBIO
nature.com/articles/s4157…
0
81
306
34,197