AlloPipe provides new insights in alloreactivity and related immune processes.
DHUYSER A. 1,2, DELAUGÈRE P. 3,4, AARNINK A. 1,2, MESNARD L. 3,5,6, RICHARD H. 3,7
1 HLA and Histocompatibility Laboratory, CHRU de Nancy, Vandoeuvre-Les-Nancy, France; 2 IMoPA 6, UMR7365 CNRS, Université de Lorraine, Vandoeuvre-Les-Nancy, France; 3 Laboratoire de biologie quantitative et computationnelle (LQCB), Institute of Computing and Data Sciences (ISCD), Sorbonne Université, Paris, France; 4 Sorbonne Université Maison des Modélisations Ingénieries et Technologies (SUMMIT), Sorbonne Université, Paris, France; 5 Soins Intensifs Néphrologiques et Rein Aigu, Hôpital Tenon, Assistance Publique des Hôpitaux de Paris, Paris, France; 6 UMR S1155, INSERM, Sorbonne Université, Paris, France; 7 Bioinformatics Units – Genome Competence Center (MF1), Robert Koch Institut, Berlin , Germany
Objectives: The historical criterion of donor/recipient (D/R) matching relies on HLA-typing. However, recipients being transplanted with a genoidentical donor may develop Graft-versus-Host-Disease after allogeneic haematopoietic stem cell transplantation or undergo acute rejection after solid organ transplantation. We and other showed that the exome comparison within a D/R pair is relevant for predicting the post-allograft outcomes. However, a gap between the quantification of exomes’ differences and the imputation of peptides capable to trigger alloreactivity – i.e. immunopeptidome – remains to fulfil.
Methods: AlloPipe is a standalone tool written in Python with few requirements and an easy installation process. It is divided into two distinct modules: Allo-Count and Allo-Affinity. Allo-Count compares exomes within two samples and returns the implied amino acid differences. Allo-Count results can then be processed with Allo-Affinity to output the immunopeptidomes’ differences within the pair.
Results: Allo-Count has been tested on a D/R cohort of kidney transplantation (n=53 pairs) showing that nsSNP quantification within the pair is correlated to graft’s chronic rejection. Allo-Affinity allows to reduce the number of D/R changes considered and outputs the list of genes and peptides candidate for priming the alloreactive response, therefore offering potential to adjust the tool and increase its power. Those later analyses are currently running, as well as the processing of the exomes of genoidentical (n=71) and haploidentical (n=40) pairs in haematopoietic stem cell transplantation.
Conclusion: AlloPipe is an open source, powerful and flexible tool that could be used to better match donor and recipient or personalize the immunosuppressive therapies according to the risk of alloreactive processes. AlloPipe analyses can also be extended to constitutional and somatic samples in context of tumoral processes, which would provide clues for the identification of specific minor histocompatibility antigens. AlloPipe is available from https://github.com/huguesrichard/Allopipe