Reference-Free Germline Immunoglobulin Allele Discovery from B Cell Receptor Sequencing Data

Antibodies, or immunoglobulins, are a diverse set of molecules that play a critical role in adaptive immunity. They are generated in a process which begins with the recombination of germline V, D, and J gene segment alleles, and refined by hypermutation of these germline sequences upon antigen exposure. Antibody repertoire analysis often requires the knowledge of the germline V, D, and J alleles to detect hypermutations and understand the phylogenetic relationships of related B cells. However, germline immunoglobulin alleles are remarkably diverse and incompletely annotated, making it necessary to construct personalized databases for every individual. Though several approaches for the detection of germline immunoglobulin variants exist, they often rely on refining existing databases using simplifying assumptions about the relationships of germline alleles in a given organism, or about the form of evolutionary process that shapes antibody repertoires. Here, we present<jats:monospace>grmlin</jats:monospace>, an alternative computational approach to detecting germline alleles. Our approach exploits two empirical properties of B cell repertoires: the abundance of germline sequences in antibody repertoires and the enormous diversity of antibody sequence space, to detect germline alleles from B cell receptor sequencing data without reliance on a reference database. As such, it is in principle applicable to non-model organisms. We validate this approach by detecting the germline alleles of 11 pairs of twins and show that it achieves equivalent sensitivity and better specificity than previous methods..

Medienart:	Preprint

Erscheinungsjahr:	2023
Erschienen:	2023

Enthalten in:	bioRxiv.org - (2023) vom: 28. Nov. Zur Gesamtaufnahme - year:2023

Sprache:	Englisch

Beteiligte Personen:	Cvijović, Ivana [VerfasserIn] Jerison, Elizabeth R. [VerfasserIn] Quake, Stephen R. [VerfasserIn]

Links:	Volltext [kostenfrei]

Themen:	570 Biology

doi:	10.1101/2023.11.25.568681

funding:
Förderinstitution / Projekttitel:

PPN (Katalog-ID):	XBI041656229