TCRalign benchmark dataset, case study dataset, and precomputed MSAs

parameter_evaluation.7z Development and independent evaluation datasets used for region-specific parameter optimization and validation of the TCR numbering algorithm. numbering_benchmark.7z Independent repertoire dataset used to evaluate IMGT numbering concordance between TCRalign and existing numbering methods. gene_assignment_benchmark.7z Simulated benchmark dataset for evaluating V/J gene assignment accuracy under different mutation levels. MSA benchmark SPTC_structures.7z Structure-based benchmark dataset used for calculating the Sum-of-Pairs (SP) and Total Column (TC) scores. The reference alignments were generated from experimentally determined TCR structures. strike_1K_1024K_dataset.7z Repertoire-scale benchmark datasets ranging from 1K to 1024K sequences, used for STRIKE score evaluation as well as runtime and peak memory benchmarking. Structure prediction tcr_fas.tar.gz Input FASTA files containing paired TCR α/β amino acid sequences. tcr_alignments.7z Precomputed multiple sequence alignments (MSAs) generated by TCRalign. tcr_pdb.tar.gz Reference experimental TCR complex structures used for structure prediction evaluation. case_dataset.tar.gz VDJServer IR-T1D-000002 case study.