Current challenges for unseen-epitope TCR interaction prediction and a new perspective derived from image classification.

The prediction of epitope recognition by T-cell receptors (TCRs) has seen many advancements in recent years, with several methods now available that can predict recognition for a specific set of epitopes. However, the generic case of evaluating all possible TCR-epitope pairs remains challenging, mainly due to the high diversity of the interacting sequences and the limited amount of currently available training data. In this work, we provide an overview of the current state of this unsolved problem. First, we examine appropriate validation strategies to accurately assess the generalization performance of generic TCR-epitope recognition models when applied to both seen and unseen epitopes. In addition, we present a novel feature representation approach, which we call ImRex (interaction map recognition). This approach is based on the pairwise combination of physicochemical properties of the individual amino acids in the CDR3 and epitope sequences, which provides a convolutional neural network with the combined representation of both sequences. Lastly, we highlight various challenges that are specific to TCR-epitope data and that can adversely affect model performance. These include the issue of selecting negative data, the imbalanced epitope distribution of curated TCR-epitope datasets and the potential exchangeability of TCR alpha and beta chains. Our results indicate that while extrapolation to unseen epitopes remains a difficult challenge, ImRex makes this feasible for a subset of epitopes that are not too dissimilar from the training data. We show that appropriate feature engineering methods and rigorous benchmark standards are required to create and validate TCR-epitope predictive models.

Leveraging T-cell receptor - epitope recognition models to disentangle unique and cross-reactive T-cell response to SARS-CoV-2 during COVID-19 progression/resolution.

Despite the general agreement on the significance of T cells during SARS-CoV-2 infection, the clinical impact of specific and cross-reactive T-cell responses remains uncertain. Understanding this aspect could provide insights for adjusting vaccines and maintaining robust long-term protection against continuously emerging variants. To characterize CD8+ T-cell response to SARS-CoV-2 epitopes unique to the virus (SC2-unique) or shared with other coronaviruses (CoV-common), we trained a large number of T-cell receptor (TCR) - epitope recognition models for MHC-I-presented SARS-CoV-2 epitopes from publicly available data. These models were then applied to longitudinal CD8+ TCR repertoires from critical and non-critical COVID-19 patients. In spite of comparable initial CoV-common TCR repertoire depth and CD8+ T-cell depletion, the temporal dynamics of SC2-unique TCRs differed depending on the disease severity. Specifically, while non-critical patients demonstrated a large and diverse SC2-unique TCR repertoire by the second week of the disease, critical patients did not. Furthermore, only non-critical patients exhibited redundancy in the CD8+ T-cell response to both groups of epitopes, SC2-unique and CoV-common. These findings indicate a valuable contribution of the SC2-unique CD8+ TCR repertoires. Therefore, a combination of specific and cross-reactive CD8+ T-cell responses may offer a stronger clinical advantage. Besides tracking the specific and cross-reactive SARS-CoV-2 CD8+ T cells in any TCR repertoire, our analytical framework can be expanded to more epitopes and assist in the assessment and monitoring of CD8+ T-cell response to other infections.

Rapid Multiplex Genotyping of 20 HLA-A*02:01 Restricted Minor Histocompatibility Antigens.

A subset of MHC-associated self-peptides presented by the recipient's cells and immunologically foreign to the donor can induce an allogeneic immune response after hematopoietic stem cell transplantation (HSCT). These immunogenic peptides originate from the genomic polymorphisms and are known as minor histocompatibility antigens (MiHA). MiHA mismatches trigger the post-transplant immune response, which could manifest in both the deleterious "graft-vs.-host" disease and the beneficial "graft-vs.-leukemia" effect. Importantly, some MiHAs are considered to be promising targets for posttransplant T-cell immunotherapy of hematopoietic malignancies. This creates a demand for a robust and fast approach to genotyping MiHA-encoding polymorphisms. We report a multiplex real-time PCR method for the genotyping of 20 polymorphisms that are encoding HLA-A*02:01-restricted MiHAs. This method uses allele-specific primers and gene-specific hydrolysis probes. In 1 h it allows for the detection of MiHA mismatches in a donor-recipient pair without the need for electrophoresis, sequencing, or other time-consuming techniques. We validated the method with Sanger and NGS sequencing and demonstrated good performance over a wide range of DNA concentrations. We propose our protocol as a fast and accurate method of identifying mismatched MiHAs. The information on the MiHA mismatches is useful for studying the allogeneic immune response following HSCT and for selecting the targets for post-transplant T-cell therapy.