Biology subjects

Huang, P.-S.

Publications and source records attributed to Huang, P.-S..

9 recordsLinked to original sources

bioRxiv2026

A generalizable system for antigenic peptide targeting across HLA-I allotypes

T cell receptors (TCRs) and TCR-mimicking antibodies recognize peptide antigens in the context of specific Human Leucocyte Antigen (HLA-I) allotypes, and the extreme polymorphism of the HLA locus limits the breadth of immunotherapy development. Key barriers include divergent molecular surfaces on HLA proteins and differences in the peptide structure. As a result, existing modalities cannot confer therapeutic coverage across patients of divergent genetic backgrounds. Here, we develop an approach which combines a peptide conformational prediction tool, PepPred, with a cross-HLA binding protein engineering system, TRACeR-I1, to outline a generalized framework for developing binders (xTRACeRs) with compatibility across HLA allotypes while maintaining high levels of specificity towards the peptide antigen. We use our system to develop and validate xTRACeRs against clinically relevant, established peptide antigens presented across common alleles within five HLA-A/B/C supertypes2. Cryo-EM structures of xTRACeR-pHLA complexes for an oncofetal antigen from PRAME and a neuroblastoma-specific peptide from PHOX2B reveal effective mechanisms to navigate polymorphic HLA surface residues, and extensive interactions with the peptide. We implemented these two xTRACeRs as Chimeric Antigen Receptor (CAR) T cells and demonstrated their potent killing efficacy and specificity. Overcoming restriction across HLA supertypes lifts a key barrier in HLA-targeted immunotherapy by expanding patient coverage.

synthetic biology↗

bioRxiv2026

De novo design of RNA pseudoknots with deep learning

RNA design has been hindered by the limited accuracy of 3D structure prediction. Here, we show that intricate RNA structures can be generated with current deep learning tools through accurate de novo design of pseudoknot secondary structures. In an Eterna competition involving 57 pseudoknots, generative AI methods matched experienced human designers in solving most blind challenges, evaluated by single-nucleotide-resolution chemical mapping, compensatory mutagenesis, and cryogenic electron microscopy. AI-generated molecules with accurate secondary structures formed well-ordered 3D folds stabilized by noncanonical tertiary interactions not modeled during design. Success was guided by an RNet foundation model trained on prior chemical mapping data, suggesting that some difficult RNA design tasks may be tractable without first solving RNA 3D structure prediction.

biophysics↗

bioRxiv2025

Assessing Generative Model Coverage of Protein Structures with SHAPES

Recent advances in generative modeling enable efficient sampling of protein structures, but their tendency to optimize for designability imposes a bias toward idealized structures at the expense of loops and other complex structural motifs critical for function. We introduce SHAPES (Structural and Hierarchical Assessment of Proteins with Embedding Similarity) to evaluate five state-of-the-art generative models of protein structures. Using structural embeddings across multiple structural hierarchies, ranging from local geometries to global protein architectures, we reveal substantial undersampling of the observed protein structure space by these models. We use Frechet Protein Distance (FPD) to quantify distributional coverage. Different models are distinct in their coverage behavior across different sampling noise scales and temperatures; the frequency of TERtiary Motifs (TERMs) further supports the observations. More robust sequence design and structure prediction methods are likely crucial in guiding the development of models with improved coverage of the designable protein space.

bioengineering↗

bioRxiv2023

Photoswitchable binders enable temporal dissection of endogenous protein function

General methods for spatiotemporal control of specific endogenous proteins would be broadly useful for probing protein function in living cells. Synthetic protein binders that bind and inhibit endogenous protein targets can be obtained from nanobodies, designed ankyrin repeat proteins (DARPins), and other small protein scaffolds, but generalizable methods to control their binding activity are lacking. Here, we report robust single-chain photoswitchable DARPins (psDARPins) for bidirectional optical control of endogenous proteins. We created topological variants of the DARPin scaffold by computer-aided design so fusion of photodissociable dimeric Dronpa (pdDronpa) results in occlusion of target binding at baseline. Cyan light induces pdDronpa dissociation to expose the binding surface (paratope), while violet light restores pdDronpa dimerization and paratope caging. Since the DARPin redesign leaves the paratope intact, the approach was easily applied to existing DARPins for GFP, ERK, and Ras, as demonstrated by relocalizing GFP-family proteins and inhibiting endogenous ERK and Ras with optical control. Finally, a Ras-targeted psDARPin was used to determine that, following EGF-activation of EGFR, Ras is required for sustained EGFR to ERK signaling. In summary, psDARPins provide a generalizable strategy for precise spatiotemporal dissection of endogenous protein function.

biochemistry↗

bioRxiv2023

An all-atom protein generative model

AO_SCPLOWBSTRACTC_SCPLOWProteins mediate their functions through chemical interactions; modeling these interactions, which are typically through sidechains, is an important need in protein design. However, constructing an all-atom generative model requires an appropriate scheme for managing the jointly continuous and discrete nature of proteins encoded in the structure and sequence. We describe an all-atom diffusion model of protein structure, Protpardelle, which instantiates a "superposition" over the possible sidechain states, and collapses it to conduct reverse diffusion for sample generation. When combined with sequence design methods, our model is able to co-design all-atom protein structure and sequence. Generated proteins are of good quality under the typical quality, diversity, and novelty metrics, and sidechains reproduce the chemical features and behavior of natural proteins. Finally, we explore the potential of our model conduct all-atom protein design and scaffold functional motifs in a backbone- and rotamer-free way.

bioengineering↗

bioRxiv2020

Ig-VAE: Generative Modeling of Immunoglobulin Proteins by Direct 3D Coordinate Generation

AO_SCPLOWBSTRACTC_SCPLOWWhile deep learning models have seen increasing applications in protein science, few have been implemented for protein backbone generation--an important task in structure-based problems such as active site and interface design. We present a new approach to building class-specific backbones, using a variational auto-encoder to directly generate the 3D coordinates of immunoglobulins. Our model is torsion- and distance-aware, learns a high-resolution embedding of the dataset, and generates novel, high-quality structures compatible with existing design tools. We show that the Ig-VAE can be used to create a computational model of a SARS-CoV2-RBD binder via latent space sampling. We further demonstrate that the models generative prior is a powerful tool for guiding computational protein design, motivating a new paradigm under which backbone design is solved as constrained optimization problem in the latent space of a generative model.

bioinformatics↗

bioRxiv2019

De novo design of proteins with two isoenergetic but structurally divergent ground states

The plasticity of naturally occurring protein structures, which can change shape considerably in response to changes in environmental conditions, is critical to biological function. While computational methods have been used to de novo design proteins that fold to a single state with a deep free energy minima (Huang et al., 2016), and to reengineer natural proteins to alter their dynamics (Davey et al., 2017) or fold (Alexander et al., 2009), the de novo design of closely related sequences which adopt well-defined, but structurally divergent structures remains an outstanding challenge. Here, we design closely related sequences (over 94% identity) that can adopt two very different homotrimeric helical bundle conformations -- one short ([~]66 [A] height) and the other long ([~]100 [A] height) -- reminiscent of the conformational transition of viral fusion proteins (Ivanovic et al., 2013; Podbilewicz, 2014; Skehel and Wiley, 2000). Crystallographic and NMR spectroscopic characterization show that both the short and long state sequences fold as designed. We sought to design bistable sequences for which both states are accessible, and obtained a single designed protein sequence that populates either the short state or the long state depending on the measurement conditions. The design of sequences which are poised to adopt two very different conformations sets the stage for creating large scale conformational switches between structurally divergent forms.

biochemistry↗

bioRxiv2019

Structure and Functional Binding Epitope of V-domain Ig Suppressor of T-cell Activation (VISTA)

V-domain Ig Suppressor of T cell Activation (VISTA) is an immune checkpoint protein that inhibits the T - cell response against cancer. Similar to PD-1 and CTLA-4, antibodies that block VISTA signaling can release the brakes of the immune system and promote tumor clearance. VISTA has an Ig-like fold, but little is known about its structure and mechanism of action. Here, we report a 1.85 [A] crystal structure of the human VISTA extracellular domain and highlight structural features that make VISTA unique among B7 family members. Through fine-epitope mapping, we also identify solvent-exposed residues that underlie binding to a clinically relevant anti-VISTA antibody. This antibody-binding region is also shown to interact with V-set and Ig domain-containing 3 (VSIG3), the recently proposed functional binding partner of VISTA. The structure and functional epitope determined here will help guide future drug development efforts against this important checkpoint target.

cancer biology↗

bioRxiv2018

Multi-Scale Structural Analysis of Proteins by Deep Semantic Segmentation

Recent advancements in computational methods have facilitated large-scale sampling of protein structures, leading to breakthroughs in protein structural prediction and enabling de novo protein design. Establishing methods to identify candidate structures that can lead to native folds or designable structures remains a challenge, since few existing metrics capture high-level structural features such as architectures, folds, and conformity to conserved structural motifs. Convolutional Neural Networks (CNNs) have been successfully used in semantic segmentation -- a subfield of image classification in which a class label is predicted for every pixel. Here, we apply semantic segmentation to protein structures as a novel strategy for fold identification and structural quality assessment. We represent protein structures as 2D -carbon distance matrices (\"contact maps\"), and train a CNN that assigns each residue in a multi-domain protein to one of 38 architecture classes designated by the CATH database. Our model performs exceptionally well, achieving a per-residue accuracy of 90.8% on the test set (95.0% average accuracy over all classes; 87.8% average within-structure accuracy). The unique aspect of our classifier is that it encodes sequence agnostic residue environments from the PDB and can assess structural quality as quantitative probabilities. We demonstrate that individual class probabilities can be used as a metric that indicates the degree to which a randomly generated structure assumes a specific fold, as well as a metric that highlights non-conformative regions of a protein belonging to a known class. These capabilities yield a powerful tool for guiding structural sampling for both structural prediction and design.\n\nSignificanceRecent computational advances have allowed researchers to predict the structure of many proteins from their amino acid sequences, as well as designing new sequences that fold into predefined structures. However, these tasks are often challenging because they require selection of a small subset of promising structural models from a large pool of stochastically generated ones. Here, we describe a novel approach to protein model selection that uses 2D image classification techniques to evaluate 3D protein models. Our method can be used to select structures based on the fold that they adopt, and can also be used to identify regions of low structural quality. These capabilities yield a powerful tool for both protein design and structure prediction.

bioinformatics↗