Bio Protocol (@BioProtocol) / X

Bio Protocol

10.3K posts

Bio Protocol

@BioProtocol

Biotech's new financial layer.

Joined May 2022

Affiliates Replies Media

Pinned
Bio Protocol
@BioProtocol
Jul 6
June in the Biosphere 🌐 • The first wet-lab resulsts for @peptai_ are in. AI-designed binders against a cancer target, tested and already feeding the next round. • BIOS ( @BioAIDevs) can now take a protein target to ranked drug candidates in under an hour, for under $7. •
Bio Protocol
@BioProtocol
Jul 6
Article
Bio Monthly (June 2026): DeSci Berlin, OpenLabs, PeptAI's First Wet-Lab Results
Highlights from June in the Biosphere 🌐 For a hundred years, getting cured meant waiting on an institution. Now the patient is the principal investigator. A founder built a cancer vaccine for his...
Bio Protocol reposted
BIOS
@BioAIDevs
5h
BIOS ranks every drug candidate using a scoring metric validated against 3,766 real lab experiments. Every cutoff is calibrated from actual binding data, so when BIOS tells you a design is promising, that confidence traces back to physical results.
00:00
Bio Protocol reposted
PeptAI
@peptai_
6h
Anyone can now rent the models that design a drug candidate. Almost no one can rent the experimental record that proves which candidates are real. Why models are commoditizing, what drug failure actually costs, and how PeptAI turns every run into a compounding dataset 🧵
00:00
Bio Protocol
@BioProtocol
Jul 12
"Yes, you should lose weight and exercise but let's cut to the chase...you should be on GLP." We sat down with Dr. Mohsen Soofian, a board-certified physician who has used peptides in clinical practice for nearly a decade. Here is what actually holds up. 🧵
00:00
Bio Protocol
@BioProtocol
Jul 12
Replying to @BioProtocol
Two areas patients ask about most: brain and sleep. For cognition, Dihexa works through the HGF pathway and drove a large jump in synaptic connections in mice. People report feeling sharper and more engaged. Because it is strongly growth-promoting, he limits it to one to two
00:00
Bio Protocol
@BioProtocol
Jul 12
The honest part: the evidence is still thin. Most protocols are extrapolated from animal models. What the field needs is large, randomized, double-blind human trials, the same bar pharmaceuticals clear. GLP-1s show what happens when that work gets done. If you are curious, his
Bio Protocol
@BioProtocol
Jul 10
Replying to @BioProtocol
DeSciNYC: Peptides 101, Tuesday July 14:
DeSciNYC: Peptides 101 · Luma
From luma.com
Bio Protocol
@BioProtocol
Jul 10
Replying to @BioProtocol
Catalyst NYC, tomorrow July 11:
Catalyst NYC: The Future of Translational Science · Luma
From luma.com
Bio Protocol
@BioProtocol
Jul 10
The Bio community is on the ground in New York this month. Two stops: Catalyst NYC by @ResearchHub (translational science, funding models, open science infra) and DeSciNYC's Peptides 101 session. If you're building in material science, enzyme engineering, microbial strain
00:00
Bio Protocol
@BioProtocol
Jul 10
OpenLabs started with a simple idea. Put the scientific method in the open, so humans and AI agents can post, challenge and develop hypotheses together. The question now is what runs on top of it. The answer, so far, is three flagship projects: a rheumatology pharmacogenomics
00:00
Bio Protocol
@BioProtocol
Jul 10
Replying to @BioProtocol
Radicular: Non-Sedating Peptide Analgesics, built with @PeptAI Lumbosacral radiculopathy, the nerve pain that runs from the lower back down the leg, is usually managed with gabapentinoids that leave patients sedated. Radicular develops drug-like peptide analogs of a published
Bio Protocol
@BioProtocol
Jul 10
All three are live on OpenLabs. Read the hypotheses, follow the agent reviews, and back the projects worth testing. Sign up now: openlabs.bio.xyz/signup
Bio Protocol reposted
Bio Protocol
@BioProtocol
Jul 8
He sold his last cancer company to Eli Lilly for $300M. His next one isn't going the traditional route. Here's the case for DeSci from someone who already won the old game 🧵
00:00
Bio Protocol reposted
PeptAI
@peptai_
Jul 8
AI can design a thousand proteins in an hour. It still cannot reliably predict which ones work, because biology never generated the data that teaches a model to generalize. A thread on the real constraint in AI drug discovery 🧵
00:00