SwiftTCR: efficient computational docking protocol of TCRpMHC-I complexes using restricted rotation matrices.

Area of Science:

• Immunology
• Structural Biology
• Computational Biology

Background:

• T cell recognition of peptides presented by MHC molecules is crucial for immune responses.
• The vast diversity of T cell receptors (TCRs) makes experimental and computational modeling of TCR-peptide-MHC (TCRpMHC) interactions challenging.
• Accurate TCRpMHC structural data is vital for developing cancer immunotherapies, transplantation strategies, and treatments for autoimmune diseases.

Purpose of the Study:

• To develop a rapid and accurate computational protocol for modeling TCRpMHC complex structures.
• To overcome the limitations of experimental methods and general-purpose docking tools for analyzing diverse TCRpMHC interactions.
• To provide structural insights into T cell recognition for therapeutic applications.

Main Methods:

• Developed an integrative modeling protocol building upon the PIPER algorithm.
• Leveraged unique docking patterns and polarized docking angles of TCRs at pMHC.
• Integrated an ultra-fast structure superimposition tool, GradPose, for accelerated clustering.
• Utilized Fast Fourier Transform (FFT) to reduce rotation sets.

Main Results:

• The protocol models TCRpMHC complexes in 3-4 minutes on 12 CPUs, achieving a 25-40x speedup compared to ClusPro.
• Outperformed state-of-the-art docking tools in model quality on a benchmark set of 38 TCRpMHC class I complexes.
• Demonstrated high computational efficiency and accuracy in predicting TCRpMHC structures.

Conclusions:

• The developed protocol offers a computationally efficient method for modeling TCRpMHC interactions.
• This approach can provide structural information for specific peptide-targeting TCR repertoires.
• Enables enrichment of TCR sequencing data and facilitates the development of structure-based deep learning algorithms for T cell recognition studies.