Recovery of TCRαβ but not TCRγδ CDR3 sequences from ancient human whole-genome sequencing data [v. 22 May 2026]

The T-cell receptor (TCR) repertoire is a major component of the adaptive immune system that reflects host-pathogen co-evolution. Previous paleogenomic studies have focused primarily on deciphering major histocompatibility complex (MHC) diversity; however, TCR repertoire diversity in ancient human populations remains largely unexplored. We investigated whether TCRαβ and TCRγδ CDR3 sequences can be recovered from ancient human whole-genome sequencing (WGS) data using TRUST4 assembly. Ten ancient WGS samples spanning approximately 35,000 years from the Upper Paleolithic to the medieval period across Eurasia and Africa were analysed; eight samples were retained following quality filtering. TCRαβ (TRA + TRB) and TCRγδ (TRG + TRD) chains were analysed separately throughout. The primary analysis was conducted on 279 productive TCRαβ sequences across eight samples, with rarefaction curves confirming near-complete sampling of the recoverable library (Good's Coverage mean 0.991 ± 0.015). In contrast, only 42 productive TCRγδ sequences were recovered (median approximately 5 per sample), which is insufficient for meaningful characterisation. TCRαβ V-gene composition showed non-random structure confirmed by PERMANOVA on Bray-Curtis distances (F = 1.544, p = 0.006). These findings demonstrate that TCRαβ CDR3 sequences are recoverable from ancient bulk WGS data across a broad range of preservation conditions, geographic locations, and chronological periods, while TCRγδ recovery is insufficient at standard WGS depth. While findings should be regarded as preliminary pending replication in larger dedicated datasets, the structured patterning observed across multiple independent analytical approaches suggests a recoverable immunological signal in ancient genomic data.