Ryan Gumpper

Assistant Professor at UNC Chapel Hill, GPCRs, Structural Biology, Musician

[Yoojoong Kim @KYoojoong]

Excited to see our recent work featured on the cover of Nature Structural & Molecular Biology!
https://t.co/1xpPIPx6dn

This is my first time being featured on a journal cover — very grateful to the NSMB team. Huge thanks, as always, to my mentor @zenbrainest !!

[Yoojoong Kim @KYoojoong]

Huge thanks to my mentor @zenbrainest for this work and for guiding me in structural biology and pharmacology. Many thanks also to @rgumpper, Yuxuan Zhuang, and Ron Dror for the MD simulations!

[Yoojoong Kim @KYoojoong]

This is out! Ever wondered why coffee is bitter? The bitter taste receptor TAS2R43 plays a central role in sensing the bitter compounds in coffee. Our TAS2R43 structures reveal how this receptor recognizes bitter molecules!

[nick_rite @nick_rite]

Congrats @KYoojoong @rgumpper @zenbrainest !

[nick_rite @nick_rite]

Yes! CryoEM is truly a single molecule experiment and we should treat the data as such!

[nick_rite @nick_rite]

It's out - we isolated mGluR2 containing assemblies from brains and resolved various conformational and compositional states. A subtype-specific functional property shapes the conformational landscape. Project like this has been a dream of mine since I started my science journey

[BryanRoth @zenbrainest]

Cryo-EM Structures of Brain-Derived G Protein-Coupled Receptors: The First Direct Visualization from Mammalian Brain Tissue https://t.co/Bzqb6tPlwJ

[BryanRoth @zenbrainest]

Cryo-EM Structures of Brain-Derived G Protein-Coupled Receptors: The First Direct Visualization from Mammalian Brain Tissue

Ka-Booom!

Wait for it soon to be on @biorxivpreprint
#gpcr

[Debora Marks @deboramarks]

Meet evedesign: a new open-source ML framework that makes protein design accessible and interoperable.

📢 See our post: https://t.co/lK3Szb3ff8

Protein design models are powerful, but combining them shouldn’t require custom glue code.

✅Combine models for multi-objective optimization
✅Integrate lab-in-the-loop experimental of data
✅100% secure: run on your own infra, no data sharing

Get started building therapeutics & industrial enzymes today 👇
📄Paper: https://t.co/eTOTY6kfZ0
💻Code: https://t.co/Hn3zatn5nc
🌐Webserver: https://t.co/cAM8NVMbL4

Reach out to collaborate: [email protected]

[Andre Cornman @ancornman1]

Predicting protein-protein interactions (PPIs) at proteome scale can take months with co-folding models due to massive all-vs-all comparisons required.

We are excited to announce FlashPPI, a contrastive model that predicts proteome wide physical interfaces in minutes. 1/🧵

[CHARMM-GUI @CharmmGui]

Support for CHARMM-GUI Infrastructure Upgrades

CHARMM-GUI serves a growing research community, and demand is pushing our current server capacity to its limits. We are raising $150,000 to upgrade our infrastructure:

https://t.co/39RHACoD95

[Biology+AI Daily @BiologyAIDaily]

A General Framework for Injecting Biophysical Priors into Protein Embeddings

1. New preprint introduces ProtBFF, an encoder-agnostic framework that enhances protein embedding models with interpretable biophysical features through cross-embedding attention.

2. The method addresses a critical flaw in SKEMPI2 benchmarking: systematic data leakage from sequence redundancy. When evaluated with proper homology-based clustering at 60% identity, existing state-of-the-art models show dramatically reduced performance.

3. ProtBFF operates as a plug-in module that reweights residue-level embeddings using five biophysical scores: interface propensity, residue burial, dihedral deviation, solvent-accessible surface area (SASA), and local distance difference test (lDDT).

4. Key innovation: rather than building specialized architectures, the framework enriches general-purpose pretrained embeddings (ESM2, ESM3, ProSST) with physical priors, enabling them to surpass dedicated binding affinity predictors.

5. Ablation studies reveal interface and burial features contribute most to accuracy, while the auxiliary ilDDT loss improves structural fidelity and generalization.

6. The 150M-parameter ESM2 with ProtBFF outperforms even 15B-parameter variants without it, demonstrating that biophysical guidance can compensate for model scale.

7. On out-of-distribution SARS-CoV-2 antibody-antigen datasets, ProtBFF-enhanced models show strong few-shot learning capability, achieving high correlation with only 10% training data.

8. The approach highlights a broader principle: pretrained representations learn rich protein features, but selective amplification through mechanistic priors unlocks task-specific performance without sacrificing generalization.

💻Code: https://t.co/WRBViwwY2n
📜Paper: https://t.co/a07zANWI7W
#ProteinDesign #MachineLearning #Biophysics #ProteinProteinInteractions #DDGPrediction #ComputationalBiology #Bioinformatics #DeepLearning

[BryanRoth @zenbrainest]

Last year @Addgene shipped 2,977 orders of @rothlab #chemogenetic viruses.

[BryanRoth @zenbrainest]

Postdoctoral fellowship available in structure guided drug discovery in @RothLabUNC
Please RT

[BryanRoth @zenbrainest]

My thoughts on psychedelic drugs and the risk of brain cancer and valvular heart diesease: https://t.co/1InausCzFg

[BryanRoth @zenbrainest]

Avoiding common errors in psychedelic drug discovery!

https://t.co/LFrfw1F9fT

[BryanRoth @zenbrainest]

Huge implications for #GPCR-based drug and sensor design
Differential Effects of Sodium on Agonist-Induced Conformational Transitions and Signaling at μ and κ Opioid Receptors | https://t.co/w9wLk96htr

[UNC Biochemistry and Biophysics department @UNC_BCBP]

UNC Neurobiology Students Explore CryoEM Technology in Hands-On Structural Biology Session
https://t.co/FtKb6UeE6G @JoshuaStrauss5 @anowlan @rgumpper @UNC_PHCO @unc_our

[Can Cao @CanCao_Natural]

I’m thrilled to share our latest work in Cell!
We developed a GoA pathway–selective 5-HT1AR agonist that may achieve rapid antidepressant effects by bypassing DRN autoreceptor feedback.
Huge thanks to my amazing team and collaborators! https://t.co/3EsgwOnQKp

[Lauren Slosky @LaurenSlosky]

Hooray!! The first major product from the lab is out today @Nature.

Small molecules that bind the GPCR-transducer interface change G protein subtype selectivity in predictable ways, enabling rational drug design.💥

Check it out! 👉https://t.co/oO21BkoxVQ

🧵👇