Mathurin Dorel

12.5K posts

Mathurin Dorel banner
Mathurin Dorel

Mathurin Dorel

@MathSRIsh

Bioinformatician, Cellular Biologist, Techbio Founder. Collecting data in a complex world @[email protected] @mathsrish.bsky.social Also #boardgames

Berlin, Germany 가입일 Eylül 2016
1.5K 팔로잉334 팔로워
고정된 트윗
Mathurin Dorel
Mathurin Dorel@MathSRIsh·
Midnight reflexion on to improve scientific evaluation beyond h-index: how about a veting mechanism with pagerank based on e.g ORCID profiles. Each researcher can vouch for any number of other researchers. The importance of a researcher is then simply their pagerank.
English
1
0
1
1.5K
Mathurin Dorel
Mathurin Dorel@MathSRIsh·
@josiezayner Any company with a sequencing service if you request it specifically. Most of them have an Aviti by now, some even have a T20.
English
0
0
0
16
Josie Zayner
Josie Zayner@josiezayner·
Looking for a full genome sequencing company that does non-human illumina PE150 and can turn around sequences in under a month anyone know anything?
English
7
2
7
2.4K
Maxx Yung
Maxx Yung@maxxyung11·
just want to put it here right now that i think openai will acquire ginkgo within 8 months
English
16
0
137
19.1K
Mathurin Dorel 리트윗함
Tom Ellis
Tom Ellis@ProfTomEllis·
Okay here's a game - you're making nanobody therapeutics. AI can help you optimise them, but you need to prioritise order of feature importance. What is your order? Affinity Specificity Expressibility Clinical toxicity In vivo stability Purified stability Patentability Other?
English
14
3
50
5.2K
Mathurin Dorel
Mathurin Dorel@MathSRIsh·
@random_poisson @EnergieDuVide_ Il faudrait mettre en place des terrains de pratique du pilotage de drones aux limites strictes. Limites enforcées par des militaires. Comme ça on entraîne les pilotes au pilotage et l'armée à la détection et la défense. Un peu comme un practice de golf.
Français
1
0
1
7
Extrême YIMBYiste
Extrême YIMBYiste@EnergieDuVide_·
Je regarde la réglementation pour savoir où je pourrais faire voler un drone et dieu que ce pays est chiant. On n'a vraiment que des vieux cons qui font les lois, on ne peut basiquement rien faire. Bonne chance pour recruter des pilotes de drone le jour où on fait la guerre.
Français
7
2
34
1.6K
Mathurin Dorel
Mathurin Dorel@MathSRIsh·
@Clint_Davey1 And it doesn't stop with the extensions. The otters play a trading game. The moles play an card engine euro. The ravens play a bluffing game. And the lizards play a zombie game as the zombies.
English
0
0
17
659
Clint Warren-Davey
Clint Warren-Davey@Clint_Davey1·
Just played Root with my game design class. It's the best way to teach them about asymmetry. You have 4 factions who are basically all playing a different game on the same board. The Vagabond is playing an RPG. The cats are playing an engine building euro. The birds are playing a wargame where you have to pre-program your turns. The Woodland Alliance are playing like the insurgents in a COIN game and spreading sympathy gradually before sparking revolts. Every time I play this with a group of students they love it. Really shows them what's possible with asymmetric wargame design.
Clint Warren-Davey tweet media
English
78
339
7K
297.8K
Mathurin Dorel
Mathurin Dorel@MathSRIsh·
@irukanji_invest Science should be. Expertise shouldn't, lack of incentive would destroy it. If you need someone to study for years to develop the required expertise, you'd better reward it or you'll run out of experts. Sometimes it means paying for models.
English
0
0
0
14
Mathurin Dorel
Mathurin Dorel@MathSRIsh·
@simonharley Naval group is one of the best industrial groups in the world. They delivered their latest nuclear submarine a year early. The PANG will be a reality by 2038. And France does have an active aircraft carrier.
English
0
0
0
100
Mathurin Dorel
Mathurin Dorel@MathSRIsh·
@p_maverick_b But yeah, we can do in vitro transcription and in vitro translation. The next step is to link both in encapsulated membranes and then decrease the number of genes in the DNA necessary to ensure proliferation.
English
0
0
1
24
Mathurin Dorel 리트윗함
Batounet de Poisson 🌐🥐
Batounet de Poisson 🌐🥐@Qraaal·
Une tendance dont je me rends compte c'est que l'IA (au sens large du terme, incluant la machine vision et le machine learning) donnent de plus en plus de valeur à de la collection de données brutes, puisqu'avant on ne pouvait pas les exploiter de façon rentable.
Français
1
2
25
1.1K
Mathurin Dorel
Mathurin Dorel@MathSRIsh·
@GergelyOrosz Don't forget the shackles of anti-trust keeping competition alive. Google is not allowed to promote Google flight and Google map beyond reasonable in their search engine. They have to worsen their product but this ensures an healthier market (done at small doses)
English
0
0
0
24
Gergely Orosz
Gergely Orosz@GergelyOrosz·
In 2010 people working at Google would have believed Google will kill most startups when they enter a category. Case in point: Google Flights. It launched in 2011. Should have killed most flight comparison websites + travel agents. Yet today they are doing better than ever…
Yuchen Jin@Yuchenj_UW

Some people at frontier AI labs told me they believe startups are over. OpenAI, Anthropic, Google, xAI will absorb every industry as AGI nears. Coding today, science, medicine, and finance next. Then everything else. If they’re right, that’s a pretty boring end of the world.

English
95
105
2.8K
476.5K
Mathurin Dorel
Mathurin Dorel@MathSRIsh·
@JacobSwett There's an FDA priority review voucher for those kind of research (Material Threat Medical Countermeasure PRV). That's a 100M+ business model. Unless it's not eligible ?
English
1
0
0
118
Jacob Swett
Jacob Swett@JacobSwett·
This nails something important: the main barriers to pathogen-agnostic defenses like far-UVC and glycol vapors are largely funding and execution shaped. If you're excited about making these technologies happen, we'd love to have you join us!
Jacob Swett tweet media
owl@owl_posting

Reasons to be pessimistic (and optimistic) on the future of biosecurity owlposting.com/p/reasons-to-b… "It was such a fun read (if you can say that about an article on weapons)!" —a glowing review from an early reader this is (once again) the longest article I have ever published at 13,000 words. it involves interviews with 16+ researchers/VC's/policy folks in this field, and discusses basically every single facet of biosecurity that i could find. topics include: how machine-learning in rapid response therapeutic design may work, the financial status of the customer base of biosecurity startups, why agroterrorism feels extremely likely to me, and a lot more i admittedly started the essay pessimistic that this subject matters at all, and i end it surprised that it doesn't keep more people awake at night. im not a doomer about it all, but i can see how people become one. very grateful to the people who decide to spend their career (or some fraction of it) working here, and especially grateful to the ones who helped teach me about the subject

English
4
6
36
10.7K
Jake Wintermute 🧬/acc
Jake Wintermute 🧬/acc@SynBio1·
Sure you could use antibiotic resistance as a selection marker to pick transformants or you could use THE TOTALITY OF SELF-REPLICATION MACHINERY THAT IS THE VERY DEFINITION OF LIFE biorxiv.org/content/10.648…
English
3
7
39
4.3K
Mathurin Dorel
Mathurin Dorel@MathSRIsh·
@_benray "reliable" is not a word I would use to describe eBay products ;)
English
1
0
0
89
Mathurin Dorel
Mathurin Dorel@MathSRIsh·
There's a good reason for that: cost. Even something as simple as a biobank can be tremendously expensive if you store it properly (aka flash frozen). FFPE is cheaper and more common but so much information is lost. And then you have the cost of measuring.
English
0
0
0
22
Mathurin Dorel
Mathurin Dorel@MathSRIsh·
We don't know how most organs develop past birth during childhood, and in general we don't know how most organs age healthily at the molecular level. Molecularly we only have snapshots of individuals at a fixed live events, rarely knowing how they got there.
Martin Borch Jensen@MartinBJensen

Very well put! AI is real. It needs data. LLMs have access to all our writing etc. Biology does not have an equivalent corpus of high-quality data that spans the dynamics we're proposing to solve. Diseases and aging occur at the level of organs and organisms, and we need data there to simulate it. Status quo won't get us there in a few years. But we can act! Identify the most important data that can't be accelerated, and start collecting it now so we can leverage AI for longevity as early as possible. We are setting up an @impetusgrants focus on AI-enabling datasets specifically.

English
1
2
7
552
ben ray
ben ray@_benray·
@MathSRIsh this is why my startup pitch cleanly explains how we build reliable integrated workcells for <$20k
English
1
0
2
588
Mathurin Dorel
Mathurin Dorel@MathSRIsh·
@_benray Opentron + 3d printer might be fun, in the vein of what @plasmidsaurus is doing. But one has to be super careful about biosafety.
English
0
0
1
15
Mathurin Dorel
Mathurin Dorel@MathSRIsh·
@_benray Fewer than 10 people have the budget for it. You need 300k+ for the robot, and then the price of the reagents. Definitely on my bucket list for our series A. Might try with an opentron in between but experiment range is more limited.
English
3
0
4
766
Mathurin Dorel
Mathurin Dorel@MathSRIsh·
@aaronmring Only a liability before a successful clinical trials, because it's an unaccounted risk factor. For drugs that passed clinical trials it's a candidate mechanism of action. It's been supposed for a while that the most effective anti cancer therapies are *because* of off target.
English
1
0
1
14
Aaron Ring
Aaron Ring@aaronmring·
The good news: off-target binding is a liability that can be fixed with clever engineering. For tanezumab (anti-NGF), which also bound TSLP, the Adimab team used directed evolution to eliminate TSLP binding while preserving (and even improving) NGF engagement.
Aaron Ring tweet media
English
4
2
15
2.8K
Aaron Ring
Aaron Ring@aaronmring·
How specific are therapeutic monoclonal antibodies, really? In our new paper, @Yile_Dai led a collaboration with Adimab to profile 174 FDA-approved and clinical-stage mAbs against 6,172 human extracellular proteins. What we found surprised us.🧵 sciencedirect.com/science/articl…
English
15
114
394
54.1K

탐색

@josiezayner @maxxyung11 @random_poisson @EnergieDuVide_ @Clint_Davey1 @irukanji_invest @simonharley @p_maverick_b