pls RT for reach! can’t wait to see y’all there 😁
@acs.org #CompChem
22.07.2025 20:47
👍 1
🔁 2
💬 0
📌 0
pls RT for reach! can’t wait to see y’all there 😁
@acs.org #CompChem
Well the NIH has cancelled the Women's Health Initiative, the largest study of women in history
It has been running continuously since 1991 and has provided massive key knowledge about diseases in women
Unreal
www.science.org/content/arti...
Register Now for Crash Course: Python For Cheminformatics-Driven Molecular Docking
This free virtual course will demonstrate the power and flexibility of python on May 1, 2025
pdb101.rcsb.org/news...
it's so close to being awesome 😭 but skimming code... ligand = any heteroatom (rip waters, ions, buffers). assumes atoms in prediction are perfectly ordered/labeled - not true for chai, boltz, etc. binding site = assumes identical residue IDs.
shameless promo, we hit most of these in np & np-bench ❤️
@fionachembot.bsky.social opened up linkedin to this, and honestly, i can't help but immediately see a mucin rendering next to this depressing news headline... 😂
People/organizations with means and a vision for the future of biomedical science - we need an “Arc Institute” equivalent in the DMV area to sponge up all the incredible talent, opportunity of a lifetime for a visionary entrepreneur, I will help! @mcuban.bsky.social
Read: OpenAI raised $40B—i.e., the entire NIH budget.
If you love drug candidates but weren't able to attend the #FirstTimeDisclosures symposium at #ACSSpring2025, check out my story about the session for @cenmag.bsky.social - with lots of cool-looking molecules.
cen.acs.org/acs-news/acs...
🚨 PLEASE RT 🚨
Are you a comp chem trainee attending Spring ACS in San Diego? Would you like to expand your network with mentors in industry & academia at the COMP Mentor Lunch?
Fill out this form if you are interested in attending!
forms.gle/Sq79TJihHnFW...
From 2010 to 2016 (latest data I have ), NIH research contributed to EVERY drug approved by the FDA
Unassigned nitrogens in nmr data often indicate biologically relevant motion in proteins, and this can be used train deep learning models of protein dynamics!
Hannah Wayment-Steele
@ginaelnesr.bsky.social @sokrypton.org
www.biorxiv.org/content/10.1...
This was a massive team effort and I'm forever grateful for the full list of authors helping me to drive this one into the endzone! Thank you to Anthony Bogetti, Carla Calvó-Tussell, @mackevinbraza.bsky.social, @lcasalino88.bsky.social, Amanda Gramm, Sean Braet, @miarosenfeld.bsky.social,... (1/2)
thrilled I got to plan this year's ACS COMP social in my science home town 🌊♥️💻🧪
hope to see y'all there!!!
This stupid SOB is practically rooting for the measles virus. He will apparently do almost anything to avoid vaccination:
reality of ML+bio research: real methodological innovation is rare, expert application is common. & that's okay! most progress comes from skillfully using existing tools. lets be honest abt contributions & stop repackaging established methods with domain expertise as "novel computational frameworks"
What's easier to learn: cryo-EM or MD simulations? Ben Orlando and I couldn't decide, so we are accepting post-doc applications from both fields! Come work with us to explore the intersection between cryo-EM and MD at Michigan State! careers.msu.edu/jobs/researc...
Mentioned in the below article, Science is running a "Trump tracker" that's attempting to keep up with the ongoing firings at scientific agencies within the US: www.science.org/content/arti...
@biophysicalsoc.bsky.social you need to be sharing at the conference this week like our lives depend on it. Because they do.
Speak up. Stand up. Get out there and protect out democracy
If anyone has the resources to resist, its HHMI. Ending this program with no notice and literally scrubbing evidence it ever existed from the website is shocking and dosappointing
i’m hiring! come work with our amazing neuralplexer development team at iambic therapeutics to help increase the conformational accuracy of our protein structure predictions!
pls share with folks you think may be interested, & feel free to reach out with questions
jobs.ashbyhq.com/iambic-thera...
this is so incredibly sweet!!! happy hanukkah ♥️ making my latkes later today and i’ll be sure to grate the onion first ;)
An example of how the limited and biased availability of data narrows the range of functional biology PLMs tap into for prediction and design
research.arcadiascience.com/pub/result-p...
it rigorously and consistently evaluates traditional structure prediction metrics like pocket-aligned ligand rmsd and dockq score. we hope this helps create more consistency in how we evaluate progress in the field of structure prediction & ultimately aids in pushing the field forward as a whole :)
also, ConfBench is just one of many aspects of the NP3 technical report - we are also excited to announce we're open-sourcing NPBench: our turnkey dataset distillation and code library for standardized structure prediction benchmarking!
unimaginably large thank you to the amazing team and leadership that made this work possible, especially Zhuoran Qiao and Matt Welborn for believing in my vision of conformational enablement of structure prediction!
& to @rommieamaro.bsky.social, who sparked my love for dynamic proteins ♡
tl;dr... this benchmark provides a quantitative framework for measuring progress in conformational prediction. we've exhausted traditional benchmarks - it's time for a paradigm shift in how we evaluate and improve protein structure prediction methods for drug discovery.
these synthetic data sources could help bridge the gap between pattern recognition and physical understanding, particularly for challenging cases like predicting full protein conformational landscapes where experimental data is scarce.
potential directions include physics-based synthetic data generation through molecular dynamics simulations, quantum mechanical calculations, and physics-informed neural networks. each approach offers unique insights into protein dynamics that complement existing structural data.
while our physics-based priors have improved performance on apo/holo predictions, we're likely reaching the limits of what can be learned from static structures alone. the field needs to consider new data sources to advance further.
there's no point in trying to get ML models to perfectly "learn" physics - we'd just end up spending the same compute as expensive physics-based models!
instead, what data signals could serve as meaningful proxies for physics-driven phenomena?