Yikes.

Each of our bimonthly HTS data releases reveals previously unknown drug/off-target interactions, but some, like this snippet from last week's Release 5, showing that capsaicin is a highly selective agonist for the orphan nuclear receptor, ESRRB, deserves special note. 🧵 1/4 🧪

🧪Excited to learn that EvE Bio's unique pharmome-mapping data has been adopted by yet another biomedical reasoning model, Stanford's Biomni (biomni.stanford.edu). Awareness of and appreciation for our efforts is really snowballing. We're grateful to be of service to the community.

🧪EvE will soon make our 1st announcement re: a potential drug repurposing opportunity open for licensing by orgs resourced to bring such things to the clinic. If your org would like early notice of such opptys, drop us a DM (or email [email protected]) w/ name, org name, & business email.

🧪 Thank-you to Abhi Mahajan for his long essay on EvE's efforts to map the pharmome. He did a good job of explaining a pretty arcane topic. "EvE really exemplifies the thesis [that] smart people in biology should do more boring things." 🤠 www.owlposting.com/p/mapping-th...

🧪 New toy! It's a robotic -20C freezer that can store 17,000 sample tubes, and upon command can spit out a rack containing just the tubes you've requested, in whatever pattern you want them. Everything is barcoded so the right sample always ends up in the right place. Or that's what they tell me.

🧪 Exciting to see EvE Bio's ever-expanding pharmome mapping dataset (data.evebio.org) cited as a source of training/test data for @futurehousesf.bsky.social's ether0 LLM, a new scientific reasoning model for chemistry. Preprint: storage.googleapis.com/aviary-publi...

🧪 EvE Bio's biggest public data release (#4) yet! 1.4k Rx drugs screened for 24 new potential off-targets including adrenergic, apelin, muscarinic, dopamine, gastrin-releasing, orexin, 5-HT, neuromedin, opioid, prostaglandin, sphingosine, somatostatin & neuromedin receptors. evebio.org/data

🧵 Today, IMPRINT, a pioneering new nonprofit initiative, is announcing $15 million in philanthropic funding to advance our work to decode the body's immune memory and uncover the causes of chronic diseases. #Immunology #Biotech #BiotechResearch

Wake up, babe. EvE Bio's data release #4 just dropped. 1.4k Rx drugs, 85 targets & counting: 237,490 combinations screened (mode/target/compound) 8 median agonists per target (potency < 1 uM) 31 median antagonists per target (potency < 1uM) 🧪Browse or DL (for non-commercial use) at evebio.org/data

Come work with us at EvE Bio in Durham, NC! EvE's Assay Dev team is looking for a pharmacology scientist to develop new cell-based screening assays. While this is a temporary position (filling in for a scientist on family leave) it has potential to become permanent. jobs.lever.co/convergentre...

Our data release #3, now live at data.evebio.org, adds 7 new nuclear receptors + 14 new 7TMs to our ever-growing list of human gene products screened against 1.4K FDA-approved drugs to ID off-targets, train AIs, repurpose drugs, and more. Enjoy! 🧪

Work with us! We seek a data scientist in the Research Triangle w/ great R & Shiny experience, comfort making technical decisions, and some scientific bkgd. This is a hybrid position requiring 2d/wk in our Durham office. 🧪https://jobs.lever.co/convergentresearch/ab999aec-2298-438a-bc54-d368b51cc6f4

🧪

EvE Data Release 2 now live! 1.4K drugs screened vs 21 7TMs: ADRA2A, ADRA2B, ADRA2C, ADRB1, ADRB2, DRD1, DRD2, GAL1, GAL2 GHSR, HRH1, HRH3, HRH4, HTR1A, HTR1B, HTR1D, HTR1F, HTR2A, HTR5, HTR6, and PTH1R in ag & antag modes. data.evebio.org For data drop announcements via email sign up at evebio.org

Thank you Schmidt Initiative For Long Covid for moving the needle on clinical trials in search of therapeutics for LC. Repurposing Evusheld for LC is an interesting long-shot. 🤞 youtu.be/Nihzqamt5xY?...

Farid's hot takes contradict many dogmas in the AI #drugdesign industry: That attaching the workhorse of the wet lab to the #AI cart will make it go; that designing small molecule drugs with AI will work at all; that all of the startups purporting to use AI are actually telling the truth. 🩺🖥️🧪💊

🧪 E11 Bio is excited to share a major step towards brain mapping at 100x lower cost, making whole-brain connectomics at human & mouse scale feasible (🧠→🔬→💻). Critical for curing brain disorders, building human-like AI systems, and even simulating human brains. Read more: e11.bio/news/roadmap

It's a very weird moment geopolitically. I'm worried for China, South Korea, France and Germany. Signs of major political instability. President Yoon Suk Yeol invoked emergency powers and declared martial law, raising fear of instability in the world’s 13th-largest economy.

Hyped about the work these guys are doing. Hopefully someone joins that can continue accelerating it. We need data!

Come work with us in Durham, NC! Are you a molecular or synthetic biologist, with mol-bio & mammalian cell culture skills, living in NC's Research Triangle region (extra points if you're experienced in drug development/discovery in academia or industry)? Let's talk! 🧪 jobs.lever.co/convergentre...

Celebrating a year of amazing efforts: by the upcoming end of its first year, EvE Bio's assay development team will have produced a whopping 200 validated HTS- and profiling-grade target actvity assays for nuclear hormone receptors (biochemical assay) and 7TM receptors (cell-based assay). Wow 😍 🧪

🧪Participants who received a single dose had a 78% lower risk of symptomatic COVID-19 infection in the first month compared to placebo. That compares quite favorably with the efficacy of approved COVID vaccines, which has been reported to be 52% at four weeks. apnews.com/press-releas...

🧪 Struggling to explain correlation? Consider using some examples from www.tylervigen.com/spurious-cor...

🧪 We asked ChatGPT4o Mini: "Does [compound X] have partial agonist activity at [gene name Y], the [receptor common name] receptor?" No one but us knows the answer to this. Like a sociopath, ChatGPT lied with a virtual grin, giving an unqualified "yes!" backed up w/ nonexistent citations. 🧵 1/3

🧪 By "the pharmome" we mean the union of all drugs, all gene products, & all the functionally relevant actions of the former on the latter. When we say "mapping the pharmome," here's what we mean, and why. We're working to make it much more than just a pipe dream. evebio.org/why-map-the-...

🧪 Why? Because you can't have a rational discussion in a Skinner box.

@science.org has arrived

🧪 www.statnews.com/2024/11/21/b...

🧪 Not an EvE Bio production, but uses our recent Data Release 1 data. Very cool to gain a peek into how others' minds approach dataviz surrounding our work. Open data unleashes enormous creativity. longitudinal.blog/2024/11/20/f...

Here's a little drawing of some of the 1400 compounds released in @evebio.bsky.social's first data dump.

Test tube tagged to include this in the Science feed. 🧪

Smart Young Folks, Hope you can make better visualizations than this notebook I whipped up colab.research.google.com/gist/adam-ma... It makes a heat-map and clusters compounds and targets (currently set to agonist mode only). (Not an official & vetted FRO work product, just me munging around.)

I have been working from a hypothesis that so far seems to be correct but I’d love to get some constructive input especially from #duckdb folks Hypothesis: Using JavaScript and DuckDB-wasm allows for the public deployment of data apps with minimal server costs and ongoing maintenance. Caveats in 🧵

We're honored to find EvE Bio's efforts cited alongside heavy hitters like Protein Data Bank and Align to Innovate in renaissancephilanthropy.org 's new playbook on philanthropic support for mid-scale science efforts in open datasets and benchmarks. Thank you!

Welcome to the Science feed! Please read our FAQs for instructions for how to be added as a contributor: bossett.io/science-feed/ Mod introductory posts linked below⬇️ Use the test tube emoji on posts you want to appear in the feed🧪 Please like the feed and make sure you follow our feed rules:

🧪 Back when we were collecting voice-of-customer for EvE Bio, some of the best advice we received was "just put all your data on the web & smart young people will instantly start analyzing the bejeebers out of it." Just released our first dataset and bejeeber #1 has already dropped! (alas, on x)

🧪 Now live: Data Release 1 of data revealing functional interactions of 1.4K small-molecule drugs with an ultimately vast # of human gene products. This 1st release is a shake-down cruise, so limited to select nuclear receptors. More to come at regular intervals. Browse, DL, enjoy! data.evebio.org

This is Fred. He just found out he can chew his own ear. There aren't actually any rules against it, he checked. 12/10

If you ever get lost in the Milky Way galaxy, here is a map.

bird sounds

🧪 In the STEM job market? Keep an eye on the positions page where our mothership, Convergent Research, lists the open positions for all its FROs like EvE. Each offers generous pay and benefits plus the opportunity to do awesome science for public good. Check it out! jobs.lever.co/convergentre...

🧪 Tyro mistake: forgot to include the test-tube emoji

Hi! We're EvE Bio, a philanthropically-funded nonprofit biomedical research institute in Durham, NC. We map the pharmome (or, at least, get a start at it...the pharmome is a big place). What does that mean and why might you care? Our CEO explains in Part I of this series: evebio.org/why-map-the-...