Deep peptide recognition profiling decodes TCR specificity and enables disease-associated antigen discovery.
T cell receptor research just got a serious upgrade. A Stanford and University of Chicago team built a new system that decodes how TCRs recognize peptides—at scale. If you’re studying immune recognition, this is the kind of tool that can change your research game.
Nat Biotechnol
by Wang N, Yeh H, Lai B et al.
“Deep peptide recognition profiling decodes TCR specificity and enables disease-associated antigen discovery. Wang N(#)(1)(2), Yeh H(#)(3)(4)(5), Lai B(3), Perera J(3), Jude KM(1)(2), Risch I(6), Um J(6), Chen X(1), Xiang X(1), Wang C(1), Liu LD(1), Yang X(7), Paley MA(6), Khan AA(8)(9), Garcia KC(10)(11)(12). Author information: (1)Department of Molecular and Cellular Physiology, Stanford University School of Medicine, Stanford, CA, USA. (2)Howard Hughes Medical Institute, Stanford University School of Medicine, Stanford, CA, USA. (3)Biohub, Chicago, IL, USA. (4)Pritzker School of Molecular Engineering, University of Chicago, Chicago, IL, USA. (5)Medical Scientist Training Program, University of Chicago, Chicago, IL, USA. (6)Rheumatology Division, Department of Medicine, Washington University School of Medicine, St Louis, MO, USA. (7)Molecular Pharmacology Program, Sloan Kettering Institute, Memorial Sloan Kettering Cancer Center, New York, NY, USA. (8)Biohub, Chicago, IL, USA. aakhan@uchicago.edu. (9)Departments of Pathology, and Family Medicine, University of Chicago, Chicago, IL, USA. aakhan@uchicago.edu. (10)Department of Molecular and Cellular Physiology, Stanford University School of Medicine, Stanford, CA, USA. kcgarcia@stanford.edu. (11)Howard Hughes Medical Institute, Stanford University School of Medicine, Stanford, CA, USA. kcgarcia@stanford.edu. (12)Department of Structural Biology, Stanford University School of Medicine, Stanford, CA, USA. kcgarcia@stanford.edu. (#)Contributed equally Predicting T cell receptor (TCR) specificity on the basis of sequence is challenging because TCRs of similar sequence can recognize entirely different antigens, whereas TCRs of different sequence can recognize the same antigens. Here we present a system that integrates high-throughput yeast display with fine-tuned protein language models (pLMs) to generate deep peptide recognition profiles (PRPs) for individual TCRs, each detailing binding against millions of peptides. We provide detailed PRPs for a panel of HLA-B*27:05-restricted TCRs from persons with ankylosing spondylitis and acute anterior uveitis that almost exclusively recognize peptides through CDR3β. pLMs trained on these PRPs outperform AlphaFold3 and tFold-TCR in predicting T cell activation. We discover and validate novel candidate autoantigens, demonstrate that model generalization to new TCRs correlates with functional distance (PRP divergence) rather than sequence similarity and introduce a model-intrinsic uncertainty metric to quantify prediction confidence. This system and its associated PRP datasets offer a scalable approach to mapping TCR recognition, accelerating antigen discovery and guiding TCR engineering. © 2026. The Author(s). Conflict of interest statement: Competing interests: The authors declare no competing interests.”
Here’s the problem: TCRs with nearly identical sequences can target totally different peptide antigens. And TCRs that look nothing alike can go after the same thing. Sequence alone doesn’t crack the code. Wang and colleagues solved this by combining high-throughput yeast display with advanced protein language models. The result? Deep peptide recognition profiles (PRPs) for individual TCRs, showing how each one interacts with millions of peptides.
Key takeaway: These PRPs don’t just map binding—they outperform AlphaFold3 and other big-name models at predicting T cell activation. That’s a major leap forward for anyone mapping TCR-antigen interactions or hunting for disease-associated antigens.
The team pushed this with a panel of HLA-B27:05-restricted TCRs from people with autoimmune conditions like ankylosing spondylitis and uveitis. They found:
PRPs reveal that these TCRs almost always recognize peptides through the CDR3β region
Their models can generalize better to new TCRs by focusing on functional similarity (PRP divergence), not just sequence
They surfaced new candidate autoantigens and introduced a built-in confidence metric for predictions
If you’re in the business of antigen discovery, TCR engineering, or just want to understand immune specificity, this approach is scalable and ready for big datasets. The peptide research index has more on methods like this for decoding peptide interactions.
This is the kind of research that moves the field forward. Expect more discoveries and tools built off these PRP datasets.
Related Reading
The STRIDE Trial and Semaglutide: Implications for Clinical Vascular Practice.
News · J Med ChemStructure-Based Adaptation of a SARS-CoV-2 Neutralizing Peptide to New Virus Variants.
News · J Nucl MedCombining an α(v)β(6)-Targeted (177)Lu-Based Peptide Receptor Radionuclide Therapy with Olaparib to Boost Therapeutic Efficacy in Pancreatic Cancer.
For Research Use Only
All content published on Pushing Peptides is intended for educational and informational purposes only. The information provided is not intended as medical advice, diagnosis, or treatment. Peptides discussed in this article are research compounds and are not approved for human therapeutic use by the FDA or any other regulatory agency. All studies referenced involve animal models or in vitro research unless otherwise stated. Consult a qualified healthcare professional before making any decisions related to your health. Pushing Peptides does not sell peptides — we are a vendor directory and educational resource.