@vasuman @markoasi1 :D okay only the Uber Eats parts left to check
English
Lukas Weidener
37 posts
No hearts were broken during the creation of this manuscript 🫀
But a few new neuronal connections were definitely made.
Big congratulations to my teammate @mihailoxyz on leveling up into a full-fledged bioinformatician through this work.
Proud of my team @AppliedSciAI
Applied Scientific Intelligence@AppliedSciAI
When most people hear a drug was withdrawn from the market, they picture something dramatic: a scandal, a coverup, a clinical disaster. The reality is usually quieter and stranger. A medicine that helped millions gets pulled because of an arrhythmia that hits a few patients out of thousands. The drug worked. It was safe for nearly everybody. But for an unlucky minority, it nudged the heart's electrical timing just enough to cause a fatal rhythm. A single ion channel - hERG - is now the leading cause of safety-related drug attrition in pharma. ~60% of new molecular entities show hERG-blocking liability in early screening. The cardiac safety filter is one of the dominant cost-of-capital decisions in early drug development. And yet the predictive tools for it have been surprisingly limited. Introducing CardioSafe CardioSafe is a multi-task neural network that predicts blocking activity across hERG, Nav1.5, Cav1.2, and IKs simultaneously - from a single chemical structure, in microseconds. CardioSafe aims to catch cardiac safety failures earlier, when they cost thousands of dollars instead of hundreds of millions, and efficiently rescue safe compounds buried in pharma archives for a fraction of the cost. The Results vs. the best published baselines on leak-free benchmarks: • AUC 0.919 vs 0.849 (CardioGenAI) and 0.819 (CToxPred2) • $51M avoided pipeline liability per 1,000-compound screen • 76% more blockers caught at equal patch-clamp budget • 39% lower cost per confirmed-safe lead in drug rescue For a mid-sized biotech running 3-5 focused libraries annually, the cumulative effect is meaningful: roughly $150 to $250M in avoided pipeline risk per year. More details in our new preprint below. Why it Works Two reasons: more data, and sharper resolution between near-identical compounds. First, CardioSafe was trained on substantially more data. Cardiac ion channel datasets are scattered across public databases in formats - censored values, inhibition-percentage votes - that prior models discarded. We kept them, with a curation policy that respects what each measurement actually means experimentally. That single choice contributes more to performance than any architectural decision we made. Second, CardioSafe was also trained to resolve activity cliffs - pairs of compounds that look nearly identical but have opposite cardiac profiles. Terfenadine (withdrawn for arrhythmias) and fexofenadine (safe, multi-billion-dollar antihistamine): same scaffold, opposite hearts. We curated 30 such pairs from the cardiac literature and fine-tuned the model to explicitly rank the blocker above its safer twin. With both molecules held out of training, CardioSafe resolves the cliff correctly. Other models flag the whole class as dangerous. The Bigger Picture CardioSafe is a proof point for how biology-native AI can run: • Multi-task prediction across structurally related targets • A closed loop with multimodal experimental assays - model proposes, MEA measures, model updates • Ruthless curation of heterogeneous public data On that last point: a @demishassabis quote recently re-surfaced on the heels of the Isomorphic raise saying the bottleneck in AI x bio isn't data - it's algorithm sophistication: "You do have enough data - if you were innovative enough on the algorithm side." Our preprint suggests a third answer. We tested multiple architectures. Cross-attention fusion, ChemBERTa embeddings, predicted transcriptomics across 978 landmark genes. They moved the headline number, but not significantly. What did: Keeping the measurements everyone else threw away and understanding what they actually mean pharmacologically. That single curation decision contributed more to performance than the architectural choices. The bottleneck isn't just more data. It isn't just better algorithms. It's also domain understanding applied to the data that already exists. These principles can extend to DILI, nephrotoxicity, neurotoxicity, and beyond. CardioSafe is the first module. The same architecture that learns to predict a drug's effect on the heart might be able to do the same in the liver, the kidney, the brain, even in plants. The platform is what we're building at ASI. Preprint & early access links below ↓
English
If your deep research is still going after an hour with 4.7, it’s either brilliant or completely derailed.
English
🤗🤗🤗introducing Hugging Science -- the home of AI for science 🤗🤗🤗
open models and datasets are the powerhouse of science (see the PDB), but finding the models and data you actually need for your breakthrough is hard af
you shouldn't need to scrape arxiv, own your own wetlab, fight a custom HDF5 parser, build a fusion stellarator, and beg for compute before you've trained a single epoch
so we're changing that
we've put all the best science on @huggingface in one place:
- 78GB of genomics data
- 11TB of PDE simulations
- 100M cell profiles
- 9T DNA base pairs
- 13M molecular trajectories
- 400k medical QA pairs
and much more, all open, and all ready for training (+ you can also now filter and search by domain, task, and keyword)
we've put together all the biggest releases from our partners at NASA, Google, OpenAI, Meta FAIR, Arc Institute, Ginkgo, SandboxAQ, Proxima Fusion, NVIDIA, Ai2, OpenADMET, InstaDeep, Future House, Polymathic AI, LeMaterial, Earth Species Project, Merck, and Eve Bio
if you're not sure where you fit in -- work on open challenges for problems that matter: including fusion stellarator design, ADMET, antibody developability, multilingual medicine, catalysis and materials, and scientific reasoning.
we're already changing how science gets done:
a fusion startup needed a benchmark for stellarator plasma confinement that didn't exist. @proximafusion shipped ConStellaration on Hugging Science: a leaderboard, dataset, and eval metrics, all in one place.
a drug discovery team wanted to predict hPXR induction. OpenADMET put up a blind challenge: 11,000+ compounds assayed at Octant, 513 held out, two tracks (pEC50 + structure). Anyone in the world can train and submit.
an antibody team at @Ginkgo released GDPa1, a developability dataset for stability, manufacturability, and immunogenicity prediction, with a live leaderboard scoring every submission.
if you know a problem the ML community should be working on, let us know. make a challenge! this is about putting all the tools for solving science in one place. so we can hillclimb!
→ huggingscience.co
English
Lukas Weidener retweetledi
Most literature agents are text-only. Biology isn't.
Alexandria, the first @AppliedSciAI agent, reads figures, tables & text like a scientist does.
• Built in collaboration with @NVIDIA
• Powered by Nemotron 3 Nano Omni
• SOTA on LitQA3, FigQA2, TableQA2
How it works 🧵
English
100%. The most powerful model isn’t one human + a swarm of agents. It’s one human operator overseeing a decentralized network of AI PIs, each running their own virtual lab.
Professor Oak@Prof_Oak_
AI+Bio mindshare has been dominated by the most ambitious, high-science projects taking on our most intractable problems Pipeline-in-a-Person flips the logic Rather than reinventing the drug discovery process, it exploits the low hanging fruit for AI - human-in-the-loop agentic orchestration - to massively parallelize the process we already follow using infrastructure that already exists. This is feasible today. Stay tuned
English
As a researcher and physician, I left the hospital to scale my impact beyond the bedside. After years of building, this is the first time I truly feel like I am close to that original vision. Proud of this team, and excited for an AI-enabled future for science.
BioAIDevs@BioAIDevs
Meet BIOS, an AI Scientist built to orchestrate complex biomedical research. • Global SOTA on Data Analysis Benchmarks: BixBench 48.78% open-answer, 55.12% multiple-choice + refusal, 64.39% multiple-choice (no refusal) - outperforming systems like Edison Scientific and Kepler. • Human-in-the-Loop or Autonomous Mode: Intermediate checkpoints let researchers guide investigations mid-flight as insights emerge. No more waiting hours for batch runs + reruns to get results. Or, run in fully autonomous mode for extended investigations. • Persistent World State: Rather than losing context as conversations grow, world state ensures investigations build on insights within each research cycle and across sessions. • Subagent Swarm: BIOS orchestrates subagents specializing in research functions (Literature Review, Data Analysis, Novelty Detection) and, soon, research domains (microbiology, longevity, genomics). BIOS is available now in Beta with free + paid tiers, exclusive launch pricing and, for limited time, free full access to academic users with a .edu email address. Pro, Researcher and Lab subscription tiers offer discounted packages on monthly credits. Our usage-based pricing is competitive and in some cases significantly cheaper than leading scientific agents. Try BIOS and read our paper in the links below ↓
English
🚀🚀🚀 New Hugging Science Project 😋😋😋
Can we predict food allergies from gut microbiome data?
A growing body of research shows strong links between microbial diversity and allergy risk, and now we’re taking the next step.
With a recent foundation model for microbiome data, we are training models to distinguish healthy and allergic profiles, paving the way for early diagnostics and prevention.
Join us as we explore how the microbiome can predict and prevent food allergies.
English
Lukas Weidener retweetledi
📑 Our article is out!
Took some time & adjustments, but now our paper on #Blockchain Literacy Testing has been published by @LedgerJrnl: ledgerjournal.org/ojs/ledger/art…
As always, many thanks & the greatest appreciation for @lukasweidener, the best research partner! 😎🤓
English
frens I introduce u to BASED @lukasweidener - he is native Web3 Doctoooooor and fierce comrade of @NootropicsDAO_ 🤝
🧠/ acc
English
English
Mainly ethics Ethan, #DeSci today is full of BS, specially on manipulated “successful” bids, like recent @QuantumBioDAO here is there “human” funders distribution:
Several suspicious patterns emerge:
1. Highly Coordinated Groups:
- 43 wallets have EXACTLY 87,005.74528539822 tokens
- 18 wallets have EXACTLY 43,502.87264269911 tokens
- 14 wallets have EXACTLY 26,101.72358561947 tokens
- 14 wallets have EXACTLY 8,700.57452853982 tokens
2. Mathematical Relationships:
- The amounts appear to be precise divisions:
* 87,005.74528539822 (base amount)
* 43,502.87264269911 (exactly 1/2 of base)
* 26,101.72358561947 (exactly 0.3 of base)
* 8,700.57452853982 (exactly 0.1 of base)
3. Suspicious Characteristics:
- These amounts maintain precision to 14 decimal places
- Perfect mathematical relationships between amounts
- Large groups of wallets with identical amounts
- Systematic distribution pattern suggesting coordinated action
This strongly suggests:
1. Automated distribution of tokens
2. Possible attempt to distribute tokens across multiple wallets while maintaining proportional control
3. Potential wash trading or artificial distribution setup
#DeSci, regardless of the sex of participants, needs ethics and accountability.
English
Hard but kinda obvious truth:
DeSci still has too much cryptoautist testosterone degenergy.
We need more estrogen and other varieties of neurodiversity in the space for it to thrive and scale.
English
@melnykowycz @FrontiersIn Excited for the final version! Was amazing working with you 🤝
English
Provisionally accepted on @frontiersin Frontiers in Blockchain "Adapting Mintzberg's Organizational Theory to DeSci: The Decentralized Science Pyramid Framework", great work with @lukasweidener. #DeSci #Web3 frontiersin.org/journals/block…
Winterthur, Switzerland 🇨🇭 English
@benjileibo where can I get a ticket for the pumped mice olympics. Want a front row seat on that
Bio Protocol@BioProtocol
English
On September 28th, I decided to stop rapamycin, ending almost 5 years of experimentation with this molecule for its longevity potential.
I have tested various rapamycin protocols including weekly (5, 6, and 10 mg dose schedules), biweekly (13 mg) and alternating weekly (6/13mg) to optimize rejuvenation and limit side effects.
Despite the immense potential from pre-clinical trials, my team and I came to the conclusion that the benefits of lifelong dosing of Rapamycin do not justify the hefty side-effects (intermittent skin/soft tissue infections, lipid abnormalities, glucose elevations, and increased resting heart rate). With no other underlying causes identified, we suspected Rapamycin, and since dosage adjustments had no effect, we decided to discontinue it entirely.
Preclinical and clinical research has indicated that prolonged rapamycin use can disrupt lipid metabolism and profiles [1], as well as induce insulin and glucose intolerance [2] as well as pancreatic Beta-cells toxicity [3]. Despite anecdotal evidence of rapamycin slowing down tumor growth, its effect in inhibiting natural killer cells [4] do raise concern for anti-cancer immune surveillance and cancer risk in the longer run.
Additionally, on October 27th, a new pre-print [5] indicated that Rapamycin was one of a handful of supposed longevity interventions to cause an increase/acceleration of aging in humans across 16 epigenetic aging clocks. This type of evaluation is the first of its kind, as most longevity interventions up to date have been tested against one or two aging clocks, leading to invisible biases and potential intended “cherry picking” of favorable clocks for the tested interventions.
Longevity research around these experimental compounds is constantly evolving, necessitating ongoing, close observation of the research and my biomarkers which my team and I do constantly.
Sources:
[1] pubmed.ncbi.nlm.nih.gov/12177161/
[2]pmc.ncbi.nlm.nih.gov/articles/PMC33….
[3]diabetesjournals.org/diabetes/artic…
[4]pmc.ncbi.nlm.nih.gov/articles/PMC40….
English
English
English
I found another fan of our device and project to onboard to @Brainstemhealth - @paulkhls founder of @Molecule_dao and @bioprotocol. I am looking forward for our interview to create ReflexDAO as part of the @bioprotocol ecosystem on Thursday.
English
