Liu Liu

All Publications

• Deep peptide recognition profiling decodes TCR specificity and enables disease-associated antigen discovery. Nature biotechnology Wang, N., Yeh, H., Lai, B., Perera, J., Jude, K. M., Risch, I., Um, J., Chen, X., Xiang, X., Wang, C., Liu, L. D., Yang, X., Paley, M. A., Khan, A. A., Garcia, K. C. 2026

  Abstract

  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.

  View details for DOI 10.1038/s41587-026-03128-x

  View details for PubMedID 42129507

  View details for PubMedCentralID 3474532

• Structural ontogeny of protein-protein interactions. Science (New York, N.Y.) Yang, A., Jiang, H., Jude, K. M., Akpinaroglu, D., Allenspach, S., Li, A. J., Bowden, J., Perez, C. P., Liu, L., Huang, P. S., Kortemme, T., Listgarten, J., Garcia, K. C. 2026; 391 (6786): eadx6931

  Abstract

  Understanding how protein binding sites evolve interactions with other proteins could hold clues to targeting "undruggable" surfaces. We used synthetic coevolution to engineer new interactions between naïve surfaces, simulating the de novo formation of protein complexes. We isolated seven distinct structural families of protein Z-domain complexes and found that synthetic complexes explore multiple shallow energy wells through ratchet-like docking modes, whereas complexes formed by natural binding sites converged in a deep energy well with a relatively fixed geometry. Epistasis analysis of a machine learning-estimated fitness landscape revealed "seed" contacts between binding partners that anchored the earliest stages of encounter complex formation. Our results suggest that "silent" surfaces have a shallower energy landscape than natural binding sites, disfavoring tight binding, likely owing to evolutionary counterselection.

  View details for DOI 10.1126/science.adx6931

  View details for PubMedID 41678610

  View details for PubMedCentralID PMC12904254

• Orientation-dependent CD45 inhibition with viral and engineered ligands Borowska, M. T., Liu, L. D., Caveney, N. A., Jude, K. M., Kim, W., Masubuchi, T., Hui, E., Majzner, R. G., Garcia, K. OXFORD UNIV PRESS. 2025

  View details for DOI 10.1093/jimmun/vkaf283.113

  View details for Web of Science ID 001627066600001

• Orientation-dependent CD45 inhibition with viral and engineered ligands. Science immunology Borowska, M. T., Liu, L. D., Caveney, N. A., Jude, K. M., Kim, W. J., Masubuchi, T., Hui, E., Majzner, R. G., Garcia, K. C. 2024; 9 (100): eadp0707

  Abstract

  CD45 is a cell surface phosphatase that shapes the T cell receptor signaling threshold but does not have a known ligand. A family of adenovirus proteins, including E3/49K, exploits CD45 to evade immunity by binding to the extracellular domain of CD45, resulting in the suppression of T cell signaling. We determined the cryo-EM structure of this complex and found that the E3/49K protein is composed of three immunoglobulin domains assembled as "beads on a string" that compel CD45 into a closely abutted dimer by cross-linking the CD45 D3 domain, leading to steric inhibition of its intracellular phosphatase activity. Inspired by the E3/49K mechanism, we engineered CD45 surrogate ligands that can fine-tune T cell activation by dimerizing CD45 into different orientations and proximities. The adenovirus E3/49K protein has taught us that, despite a lack of a known ligand, CD45 activity can be modulated by extracellular dimerizing ligands that perturb its phosphatase activity and alter T cell responses.

  View details for DOI 10.1126/sciimmunol.adp0707

  View details for PubMedID 39454026