David Ryan Koes
Removing barriers to computational drug discovery one bit at a time. Associate Professor in Computational and Systems Biology at the University of Pittsburgh.
bits.csb.pitt.edu
- Had an informative and enjoyable time at NeurIPS and MLSB. It was great to see my students present their work and catch up with other CPCB students, both past and present.
- Excited to be at the first independent @workshopmlsb.bsky.social
- If you like to sample from the Boltzmann distribution and are in San Diego for NeurIPS, be sure to check out Rishal's (@rishalchich.bsky.social) poster (#2110). Great work with Nick Boffi (@nmboffi.bsky.social) and Jacky Chen. neurips.cc/virtual/2025... arxiv.org/abs/2507.00846
- Reposted by David Ryan Koes🚨To accommodate the addition of EuroMLSB, we have extended the submission deadline to October 1, 2025 11:59pm AoE. Find information on paper guidelines at mlsb.io. Submissions will be made through CMT.
- 📢The submission portal for MLSB 2025 is live! Submissions are due September 26, 2025 11:59 pm AoE. cmt3.research.microsoft.com/MLSB2025
- Reposted by David Ryan KoesMost of MLIPs dont distinguish between spin states, making them unsuitable for open-shell chemistry. We present AIMNet2-NSE (Neural Spin-charge Equilibration), MLIP that incorporates spin-charge equilibration for systems with arbitrary charge and spin. #compchem #skychem chemrxiv.org/engage/chemr...
- Woohoo!
- The MLSB workshop will be in San Diego, CA (co-located with NeurIPS) this year for its 6th edition in December 🧬🔬 Stay tuned @workshopmlsb.bsky.social as we share details about the stellar lineup of speakers, the official call for papers, and other announcements!🌟
- Reposted by David Ryan KoesProtein function often depends on protein dynamics. To design proteins that function like natural ones, how do we predict their dynamics? @hkws.bsky.social and I are thrilled to share the first big, experimental datasets on protein dynamics and our new model: Dyna-1! 🧵
- Nooooooooo.... 😲
- Unfortunately, the MLSB Workshop @ NeurIPS (@workshopmlsb.bsky.social) was rejected this year. Feedback from the deciding committee indicates it was a coin flip decision, with 283 proposals & a number related to “computational biology” More on the future of MLSB soon…
- Reposted by David Ryan KoesNIH funding supporting the HMMER and Infernal software projects has been terminated. NIH states that our work, as well as all other federally funded research at Harvard, is of no benefit to the US.
- Reposted by David Ryan KoesOur new preprint PharmacoForge: Pharmacophore Generation with Diffusion Models is out now! PharmacoForge quickly generates pharmacophores for a given protein pocket that identify key binding features and find useful compounds in a pharmacophore search. Check it out! 🧪 doi.org/10.26434/che...
- Was proud and honored to hood Dr. Drew McNutt at the Pitt School of Medicine Diploma Ceremony. Here we are rocking both the blue and gold and tartan colors representing our joint Pitt-CMU CompBio PhD program. Congratulations to Drew and the other @cmupittcompbio.bsky.social graduates!
- Reposted by David Ryan KoesI am #exhilarated to share that our new paper "LIGYSIS-web: a resource for the analysis of protein-ligand binding sites" is now published in @narjournal.bsky.social! After almost two years of development and more than 600 commits, LIGYSIS-web is out! 📜 : tinyurl.com/utges-LIGYSI...
- Reposted by David Ryan KoesTwo students in the Joint Carnegie Mellon-University of Pittsburgh PhD Program in Computational Biology have received honorable mentions from the National Science Foundation’s Graduate Research Fellowship Program. Congratulations to Anamarie Martinez and Emma Flynn! Read more: tinyurl.com/GRFPPitt
- Reposted by David Ryan KoesWe're recruiting a 3-year postdoc for the Novo Nordisk - Oxford Fellowship programme! Develop machine learning approaches for fragment library design and experimental optimisation With @fergusimrie.bsky.social and Charlotte Deane Job advert: shorturl.at/3l47e Further details: shorturl.at/u4UkK
- Everything is easy in 2D.. - @franknoe.bsky.social
- New "blogpost" from our lab, that got accepted at ICLR 2025! We compare an old MCMC method known as Sequential Monte Carlo to generative models trained on energy functions (iDEM/iEFM) and show that MCMC does better. Check it out here: rishalaggarwal.github.io/ebmvsmcmc/
- Reposted by David Ryan Koes🚀One week left to register for our Symposium on Open Drug Discovery! Join us in Montreal April 7-8 for an exciting program showcasing how open science and AI are driving drug discovery. Some sessions are already sold out, so register now! Register by April 2nd: conscience.ca/symposium2025
- Interested in generative modeling and pharmacophores search for SBDD? Check out our talks at #ACSSpring2025
- Reposted by David Ryan Koes
- Reposted by David Ryan KoesA pie graph worth keeping in mind as the NIH budget plummets jamanetwork.com/journals/jam... for 356 new FDA drugs approved
- Reposted by David Ryan KoesOur latest ChemRxiv preprint, "Transferable #MachineLearning Interatomic Potential for Pd-Catalyzed Cross-Coupling Reactions" Collaboration with @nsf-ccas.bsky.social @gabegomes.bsky.social @bobbypaton.bsky.social chemrxiv.org/engage/chemr... #compchem #chemsky
- Reposted by David Ryan KoesThe CACHE #2 preprint is now online: bit.ly/3DyCmHN Active learning and fragment growing delivered confirmed hits. A citizen scientist using the Fold-it gaming interface designed the top compound! Kudos to Sasha and Madhushika @thesgc.bsky.social at the bench. @conscience-network.bsky.social
- Reposted by David Ryan KoesAndrew McNutt, '24 CPCB graduate, is ready for life's next adventure. Before he begins his career in drug discovery, he's taking a well-deserved break to hike the Appalachian Trail from Georgie to Maine. Read more: tinyurl.com/AndrewMcNutt
- Reposted by David Ryan KoesWelcome to the Bluesky account for Stand Up for Science 2025! Keep an eye on this space for updates, event information, and ways to get involved. We can't wait to see everyone #standupforscience2025 on March 7th, both in DC and locations nationwide! #scienceforall #sciencenotsilence
- Reposted by David Ryan KoesAlphaFold can predict protein, DNA, or RNA complexes.. BUT is it easy to determine the reliable interactions?🤔 ➡️ AlphaBridge will identify the most confident interactions, making your analysis easier! 🌐Try it here: alpha-bridge.eu 👀Read more: www.biorxiv.org/content/10.1...
- Reposted by David Ryan KoesNIH council meeting (NIDCR)scheduled for today, where our R01 grant (3rd percentile) was supposed to be discussed, was postponed indefinitely as a consequence of Trump's executive order which ordered pausing of all communications from all federal agencies. Not good.
- Reposted by David Ryan KoesProtein backbone diffusion models consistently undersample catalytically important motifs, and oversample idealized helices, raising questions about the appropriateness of these methods in designing starting points for enzyme design & evolution. From www.biorxiv.org/content/10.1...
- Reposted by David Ryan KoesOPIG is now on Bluesky! Follow us for updates about the group's latest work, web app updates, and more. opig.stats.ox.ac.uk
- Reposted by David Ryan KoesIf you want to learn more about BioEmu, join Sarah Lewis and Michael Gastegger in @hannes-stark.bsky.social ‘s online seminar today. bsky.app/profile/hann...
- Super excited to preprint our work on developing a Biomolecular Emulator (BioEmu): Scalable emulation of protein equilibrium ensembles with generative deep learning from @msftresearch.bsky.social ch AI for Science. www.biorxiv.org/content/10.1...
- Rishal’s colab notebook will suggest pharmacophores given a receptor structure that can be directly imported into Pharmit for screening. Check it out!
- New paper at the turn of the new year :) Check out our latest paper on using reinforcement learning for pharmacophore elucidation on protein binding sites. It comes with a google colab notebook for use. bmcbiol.biomedcentral.com/articles/10....
- Reposted by David Ryan KoesI wrote up a 2024 Year in Review for @avogadro.cc 🧪⚗️ #opensource #chemsky In short, the community is amazing. Significant translations into 17 languages including 🇭🇷🇫🇷🇭🇺🇰🇷🇷🇴🇪🇸Tamil, 🇺🇦🇨🇳🇬🇪🇩🇪🇯🇵🇵🇹🇧🇷🇷🇸🇹🇷 .. and of course dozens of features and bug fixes discuss.avogadro.cc/t/2024-year-...
- These ligands make up 33% of the v2 plinder test set (shown with counts): MG 94 NAG 72 NAG-NAG 64 ZN 44 MN 32 BMP 30 HEM 24 EDO 18 SO4 17 PO4 16 ADP 15 GOL 13 ANP 13 CA 13 FMN 11
- Here's a year-end update from #PLINDER. It's been really great working on this project and all the other projects that it has kickstarted - the gift that keeps on giving. Happy holidays everyone! www.plinder.sh/blog/updates
- Reposted by David Ryan KoesHere's a year-end update from #PLINDER. It's been really great working on this project and all the other projects that it has kickstarted - the gift that keeps on giving. Happy holidays everyone! www.plinder.sh/blog/updates
- Reposted by David Ryan KoesBoltz v0.4.0 is here! Today, we’re releasing our full data processing pipeline, making it easier than ever to build on top of Boltz. This release also includes our evaluation code and new results. Oh, and also pocket conditioning :)
- Reposted by David Ryan KoesHi Bluesky, just in time for the holidays I am excited to share the latest pre-print from my group! We solved the 3D structure of a mysterious viral RNA that resists degradation by host nucleases. A short 🧵 &link below – please also check out the full video! #RNA #RNAbiology #RNASky #lovevirology
- Reposted by David Ryan KoesLast week several of our students presented their research at NeurIPS 2024 in Vancouver. We are so proud of the excellent work that they've accomplished and shared with the scientific community! 🧪 Check out each of their papers below:
- Reposted by David Ryan KoesNew version of Dockformer preprint landed on arXiv. This virtual screening docking method outperforms all Deep Learning-based approaches, achieving success rates of 90.53% on PDBbind and 82.71% on PoseBusters. Expected to accelerate drug discovery and reduce drug design costs.
- Reposted by David Ryan Koes[This post could not be retrieved]
- Reposted by David Ryan KoesWith @polarishub.io we’ve built out infrastructure to make hosting and using large scale, ML ready dataset and benchmarks easy. In addition to the dataloaders, we could also maintain leaderboards. Would be happy to explore ways in which we can support the awesome work you’re doing with Plinder.
- Reposted by David Ryan KoesThe American Journal of Human Genetics sat with @fritzroth.bsky.social in the latest "Inside AJHG" to discuss his recently published paper, “A missense variant effect map for the human tumor suppressor protein CHK2.”➡️https://www.ashg.org/ajhg/inside-ajhg-with-fritz-roth/ #ASHG #GeneticsDiscoveries
- Going into @workshopmlsb.bsky.social , a few thoughts on the PLINDER dataset. This is a monumental and highly appreciated undertaking for extracting protein-ligand interactions from the PDB. First the good...
- A critical part of evaluating the generalizability of models is having a robust test set that does not have leakage from the training set (you shouldn't be able to get the correct answer by memorizing the training set).
- PLINDER has a single test set that is separated from the training set by removing all systems that are too similar to the test set. We've been looking at descriptors of pocket similarity that are distinct to those used in the construction of the PLINDER test/train split.
-
View full threadUsing PLINDER train/test uncritically for protein-ligand training is unlikely to end well (but easy enough to fix with stratification and quality filtering).
- Reposted by David Ryan KoesI'm presenting a new paper "Exploring Discrete Flow Matching for 3D De Novo Molecule Generation" at @workshopmlsb.bsky.social this week! More info in this thread but reach out if want to chat at NeurIPS about generative models or molecular design. arxiv.org/abs/2411.16644
- Don’t let the standing room only discourage you from checking out @ian-dunn.bsky.social and Drew McNutt’s posters!
- Reposted by David Ryan KoesConcept Bottleneck Language Models For protein design - Introduce CB-pLM (Concept Bottleneck Protein Language Models) from 24M to 3B, trained on UniRef50 and SwissProt over 718 concepts (including Cluster name, Biological process, and Biopython-derived features, etc.) arxiv.org/abs/2411.06090
- It’s worth opening the notebook if only to see the cool molecular animations in py3dmol.
- FlowMol at your fingertips! We just released a colab notebook to make using FlowMol super easy. Come chat with us tomorrow at @workshopmlsb ! #NeurIPS2024 🧪 colab.research.google.com/github/Dunni...
- Reposted by David Ryan KoesFlowMol at your fingertips! We just released a colab notebook to make using FlowMol super easy. Come chat with us tomorrow at @workshopmlsb ! #NeurIPS2024 🧪 colab.research.google.com/github/Dunni...
- I'm presenting a new paper "Exploring Discrete Flow Matching for 3D De Novo Molecule Generation" at @workshopmlsb.bsky.social this week! More info in this thread but reach out if want to chat at NeurIPS about generative models or molecular design. arxiv.org/abs/2411.16644
- Reposted by David Ryan KoesEarly December, the CASP16 Meeting was held in the Palladium resort in Punta Cana, Dominican Republic. Nick Grishin, myself, and Qian Qong presented the assessment of the monomeric structure predictions, the assembly predictions and of monomers in their oligomeric context. 1/n
- Reposted by David Ryan KoesExcited about BioEmu? Opening this position for just a few days over #NeurIPS2024. Looking especially for Bioinformatics + Structural Biology skills, MD/Stat Mech skills and/or #deeplearning architecture design + engineering skills aka.ms/ai4science-r...
- Super excited to preprint our work on developing a Biomolecular Emulator (BioEmu): Scalable emulation of protein equilibrium ensembles with generative deep learning from @msftresearch.bsky.social ch AI for Science. www.biorxiv.org/content/10.1...
- More excellent work from the indomitable @ian-dunn.bsky.social Come check it out at @workshopmlsb.bsky.social
- I'm presenting a new paper "Exploring Discrete Flow Matching for 3D De Novo Molecule Generation" at @workshopmlsb.bsky.social this week! More info in this thread but reach out if want to chat at NeurIPS about generative models or molecular design. arxiv.org/abs/2411.16644
