Peter E Cavanagh

Excited to share this work on biasing protein conformational occupancy, now out in @ScienceMagazine, co-led with Andrew Xue. Check out Alice's thread, and video (x.com/aliceyting/sta…) I want to expand on a few things we learned along the way.. 1/9

Next Tues (3/3) at 4PM ET, @PeterECavanagh and Andrew Xue will present "Computational design of conformation-biasing mutations to alter protein functions" Paper: science.org/doi/10.1126/sc… Then, on Wed (3/4) at 4PM ET, @sarahgurev will give an early-career talk!

Thank you! And great questions. I think CB should be used for selecting mutations that are likely to shift conformational occupancy, and CB will therefore allow researchers to hone in on variants of interest quickly. However, variants can have profound effects without conformational biasing, and CB has low recall. Also important to note that CB requires accurate structures of relevant conformational states. Basically, I think it will be useful under many circumstances but probably not a standard step. Regarding the 'tunability' - this seems to be consistent across datasets: biasing towards an active conformation results in increased basal activity of the enzymes we've looked at.

@PeterECavanagh @ScienceMagazine Great job and congrats! Do you see this becoming a standard pre-filter step before running heavy MD simulations for variant discovery? Also did this 'tunability' hold up across other flexible enzymes you tested?"

Excited to share this work on biasing protein conformational occupancy, now out in @ScienceMagazine, co-led with Andrew Xue. Check out Alice's thread, and video (x.com/aliceyting/sta…) I want to expand on a few things we learned along the way.. 1/9

I'm incredibly grateful to have worked with such an exceptional team: Andrew Xue, @ShizhongD, Albert Qiang, Tsutomu Matsui, and @aliceyting. Check out our paper here: science.org/doi/10.1126/sc… 9/9

Limitations In the paper, we discuss a number of limitations of CB, but an important one is that CB has low recall. It is quite good for protein design: the mutations CB predicts with the highest biasing scores are highly enriched for the desired conformational effect. But if you want to know the conformational effect of your variant of interest, CB simply may not pick up the signal. Andrew made CB easily accessible: (github.com/alicetinglab/C…) Just note, the structures you input must be relevant conformational states of your protein. CB works with AlphaFold-predicted structures, and with variants >1 mutation from the wild type sequence, but there will definitely be limitations here as well. 8/9

Experimental validation We validated CB with 6 more DMS datasets and then applied this method to the E. coli enzyme Lipoic Acid Ligase (LplA). We wanted to test whether CB-predicted variants of LplA were biasing the conformational equilibrium by more directly measuring conformational occupancy, so we teamed up with Tsutomu Matsui from SLAC to use SEC-SAXS to study LplA’s conformational state. As an orthogonal method, we used a tryptophan intrinsic fluorescence assay to also measure conformational occupancy of LplA variants. Trp-fluorescence and SAXS showed good agreement (see figure below) and indicated that the selected biasing mutations indeed shifted LplA towards the intended conformational state. “Neutral”-predicted double mutants (purple), made by combining “open” (red) and “closed” (blue) mutations, were measured to be intermediate in their conformational state.

Design with specificity We first validated CB on 7 Deep Mutational Scanning (DMS) datasets of proteins with conformation-specific functionality. One example is the small GTPase K-Ras, for which Weng et al. (nature.com/articles/s4158…) experimentally evaluated >30,000 single and double variants for binding to a panel of state-specific K-Ras binders. This dataset included binders to both inactive and active states of K-Ras, allowing us to test whether CB indeed is tuning protein conformation with specificity towards each desired conformational state. 3/9

Excited to share the first paper from my lab in @Nature ! We repurposed the widely used research tool, proximity labeling, to demonstrate its therapeutic potential in antigen amplification for immunotherapy using mouse tumor models and patient samples. nature.com/articles/s4158…

@WeiWei_Stanford @UWBiochem @uwmadisonipib Congrats Wei!! :)

I'm beyond thrilled to announce that I will be joining the Biochemistry Department at the University of Wisconsin–Madison this fall as an Assistant Professor. My lab will focus on deorphanizing biochemical pathways that regulate nutrient metabolism @UWBiochem @uwmadisonipib

So excited to share our work developing a new class of modular synthetic GPCRs, out today @Nature! Check it out nature.com/articles/s4158… Synthetic receptors that mediate antigen-dependent cell responses are transforming therapeutics, drug discovery and basic research. However, established technologies such as chimeric antigen receptors can only detect immobilized antigens, have limited output scope and lack built-in drug control. Here we engineer synthetic G-protein-coupled receptors (GPCRs) that are capable of driving a wide range of native or non-native cellular processes in response to a user-defined antigen. We achieve modular antigen gating by engineering and fusing a conditional auto-inhibitory domain onto GPCR scaffolds. Antigen binding to a fused nanobody relieves auto-inhibition and enables receptor activation by drug, thus generating programmable antigen-gated G-protein-coupled engineered receptors (PAGERs). We create PAGERs that are responsive to more than a dozen biologically and therapeutically important soluble and cell-surface antigens in a single step from corresponding nanobody binders. Different PAGER scaffolds allow antigen binding to drive transgene expression, real-time fluorescence or endogenous G-protein activation, enabling control of diverse cellular functions. We demonstrate multiple applications of PAGER, including induction of T cell migration along a soluble antigen gradient, control of macrophage differentiation, secretion of therapeutic antibodies and inhibition of neuronal activity in mouse brain slices. Owing to its modular design and generalizability, we expect PAGERs to have broad utility in discovery and translational science. Huge thanks to my advisor and mentor @aliceyting, fellow Ting lab members and co-authors @ReikaTei, @MatthewRavalin, Bo Cai, and of course, our amazing collaborators @yulonglilab, Yuqi Yan, @SolteszLab, and Peter Klein.

And it’s out! Our lab’s very first paper is published today @naturemethods! Our calcium-activated split-TurboID (CaST) can biochemically tag activated neurons within 10 minutes, enabling a molecular handle on functionally relevant cell ensembles: tinyurl.com/ddyecxze

We're proud to announce #BaseFold, an #AI model that improves 3D protein structure prediction by up to 6X and small mol docking by up to 3X when compared to #AlphaFold2 - DM us to see how we can accelerate your #drugdiscovery / #proteinengineering efforts: pharmaphorum.com/news/basecamp-…