Liu (Liao) Family Professor of Bioengineering, ChEM-H @Stanford.
Stanford, CA
- Tweets 455
- Following 646
- Followers 5,853
- Likes 11,063
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
60
338
30,150
Michael Fischbach retweeted
Genomes are full of dark matter of unknown functions.
We present Minerva, a new method for discovery guided by genome language models. Minerva reveals the interactions hidden in non-coding DNA, pointing to hundreds of new putative RNAs and repetitive elements per bacterial genome. Minerva allows us to find and study elements invisible to traditional methods at orders of magnitude greater scale than before.
13
150
581
41,011
Michael Fischbach retweeted
Excited to share Minerva, our approach using genome language models for biological discovery! Using Minerva, we find that UG27 reverse transcriptase systems encode variable arrays of diverse ncRNAs with a shared structure, each templating a short DNA hairpin. With @garykbrixi.
20
110
458
113,213
Michael Fischbach retweeted
The U.S. FDA approved the first multi‑selective oral RAS inhibitor in metastatic pancreatic cancer.
More: revmed.com/media/news/13366/…
17
53
183
22,088
A motivated team in biotech can work miracles.
Our story:
FDA approves new pancreatic cancer drug expected to usher in new era of treatment
The therapy, Rasonque from Revolution Medicines $RVMD, nearly doubled overall survival in a key clinical trial
statnews.com/2026/08/26/panc…
1
26
4,150
Finally, best news of the morning (and something you didn't expect so soon)...
Revolution Medicines $RVMD daraxonrasib Ph3 results in second-line pancreatic cancer.
Median overall survival in ITT patients (KRAS mutants + wild type combined) Dara 13.2 months vs chemo 6.7 months.
OS Hazard ratio 0.40
4
8
71
10,683
Michael Fischbach retweeted
I am so excited to share our new paper in @Nature: the first programmable, site-specific integration of a large DNA payload into T cells in vivo.
A single IV injection results in therapeutic levels of TRAC-targeted CAR T cells in multiple models.
nature.com/articles/s41586-0…
a 🧵
21
140
547
33,486
Wow -- I hadn't seen Google announce this capability. I strongly encourage you try this set of prompts in Gemini:
1. Can you read my email?
2. Excellent. Are you able to see my calendar as well?
1/2
3
2
50
32,964
3. Terrific. Do you have access to my Google docs? I'm wondering specifically about a doc called ***, and another called ***.
4. Wow, this is great. A few questions for you. First, are you able to propose email replies *in my style*, by reading and emulating my word choice and sentence structure, etc?
5. Second, by reading through my emails, all of which are stored my Gmail account (I never delete emails), can you discern patterns in how I respond to 'retail' emails in which people are asking to meet? Can you propose replies to people I typically agree to meet with offering times that are free in my calendar? And can you gently decline (or propose that I simply don't respond) to people to whom I typically don't offer meeting times?
2/2
1
1
13
6,386
Michael Fischbach retweeted
Proud supporters of this at @open_phil after we saw @mfgrp’s wild paper on it a few years ago. Much left to figure out!
Chasing down “big-if-true” findings like this is a good fit for science philanthropy, since donors can move quickly, flexibly, reactively vs govt grant cycles
Saw a talk this week on the trippiest thing: bacterial cultures on skin induce immune responses. At Michael Fischbach's lab, they demonstrated immunizing mice against tetanus by just dabbing an engineered version of a harmless staph culture on its head. Unlike intramuscular vaccination, the antibody levels don't seem to demonstrate any waning, even after a year.
This is both scientifically very interesting (can vaccines just be... creams?), and an amazing reminder of how much we still don't know. We could have discovered this decades ago, but somehow never noticed until now.
2
12
151
15,305
Michael Fischbach retweeted
Saw a talk this week on the trippiest thing: bacterial cultures on skin induce immune responses. At Michael Fischbach's lab, they demonstrated immunizing mice against tetanus by just dabbing an engineered version of a harmless staph culture on its head. Unlike intramuscular vaccination, the antibody levels don't seem to demonstrate any waning, even after a year.
This is both scientifically very interesting (can vaccines just be... creams?), and an amazing reminder of how much we still don't know. We could have discovered this decades ago, but somehow never noticed until now.
64
234
2,524
231,451
Michael Fischbach retweeted
The Weill Cancer Hub is fostering groundbreaking @UCSF–@Stanford collaborations to develop transformative cancer therapies. I’m grateful to co-lead one of the selected teams with @mfgrp to advance next-generation in vivo CAR T cell therapy!
Backed by a $100M gift from Joan and Sandy Weill, UCSF and @StanfordMed @StanfordCancer are launching Weill Cancer Hub West, a $200M initiative in team science to accelerate cancer research and improve care over the next decade.
tiny.ucsf.edu/ppVRqt
2
4
77
7,086
Michael Fischbach retweeted
I’ve been busy this week… 👀
Passed my PhD defense!
Insanely grateful for @mfgrp. Truly epic scientist and human.
33
2
206
19,274
Michael Fischbach retweeted
What an amazing story that was such a blast to collaborate on with @mfgrp lab!!! Phenomenal work led by @DjenetBousbaine and all co-authors!! Unroll this thread and be amazed 🤯👇🏾
Today we report that an engineered skin bacterium, swabbed gently on the head of a mouse, can unleash a potent antibody response against a pathogen. Could lead to topical vaccines that are applied in a cream. @DjenetBousbaine led the charge... @Nature 1/55
2
6
36
9,096
Michael Fischbach retweeted
A typical project for an incoming graduate student might involve 1–2 weeks of planning and 2–5 years of execution, [yet] the problem you choose will influence the impact of your work just as much as the quality of your execution. – Michael Fischbach
buzzsprout.com/1744020/episo… @mfgrp
3
21
113
18,436
Michael Fischbach retweeted
@SU2C Scientific Advisory Committee member @mfgrp published a paper in @Nature showing that immunity to a common skin bacteria involves a coordinated T and B cell response, which can be redirected against pathogens as a novel form of topical vaccination nature.com/articles/s41586-0…
1
7
1,318
Michael Fischbach retweeted
Extremely proud to find this innovative vaccine platform from Michael Fischbach at Stanford. Imagine no more jabs! Low cost and globally deployable. thetimes.com/uk/science/arti…
1
11
1,749
Michael Fischbach retweeted
The skin — once thought to be a mainly passive barrier — can produce its own antibodies that fight off infections
nature.com/articles/d41586-0…
4
72
249
32,326
Michael Fischbach retweeted
take some time now to read this thread
it explains what happens with the immune system when swabbing (no entry!) a bacteria on the head of a mouse
it beautifully walks us through the years-long scientific journey
and is exemplifies how our competitors are in fact our friends❤️
Today we report that an engineered skin bacterium, swabbed gently on the head of a mouse, can unleash a potent antibody response against a pathogen. Could lead to topical vaccines that are applied in a cream. @DjenetBousbaine led the charge... @Nature 1/55
2
18
2,983
Michael Fischbach retweeted
imo this is one of the most exciting lines of immunology research in recent years
Today we report that an engineered skin bacterium, swabbed gently on the head of a mouse, can unleash a potent antibody response against a pathogen. Could lead to topical vaccines that are applied in a cream. @DjenetBousbaine led the charge... @Nature 1/55
1
11
144
17,028