Identification of subject-specific immunoglobulin alleles from expressed repertoire sequencing data

Gadala-Maria, Daniel; Gidoni, Moriah; Marquez, Susanna; Vanderheiden, Jason Anthony; Kos, Justin T.; Watson, Corey T.; O’Connor, Kevin; Yaari, Gur; Kleinstein, Steven H.

doi:10.1101/405704

Cited by 22 publications

(36 citation statements)

References 33 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…When such mismatches are repeatedly seen in a large set of VDJ rearrangements involving multiple IGHJ genes, having diverse CDR3 regions and amplified from a single individual, the inference of a previously undiscovered gene polymorphism may be made with confidence. The discovery of allelic variants by inference is now a feature of most human repertoire studies, and this is facilitated by a number of recently developed utilities 24,25,26,27 (https://arxiv.org/abs/1711.05843). When this approach was applied to the mouse, the outcome was quite unexpected.…”

Section: Introductionmentioning

confidence: 99%

A comparison of immunoglobulin IGHV, IGHD and IGHJ genes in wild-derived and classical inbred mouse strains

Watson

Kos

Gibson

et al. 2019

Preprint

Self Cite

View full text Add to dashboard Cite

To better understand the subspecies origin of antibody genes in classical inbred mouse strains, the IGH gene loci of four wild-derived mouse strains were explored by analysis of VDJ gene rearrangements. A total of 341 unique IGHV gene sequences were inferred in the wild-derived strains, including 247 sequences that have not previously been reported. The genes of the Non-Obese Diabetic (NOD) strain were also documented, and all but one of the 84 inferred NOD IGHV genes have previously been observed in C57BL/6 mice. This is surprising because the Swiss mouse-derived NOD strain and the C57BL/6 strain have no known shared ancestry. The relationships between the genes of the wild-derived inbred strains and of the C57BL/6, NOD and BALB/c classical inbred strain were then explored. The IGH loci of the C57BL/6 and the MSM/MsJ strains share many sequences, but analysis showed that few sequences are shared with wild-derived strains representing the three major subspecies of the house mouse. There were also few IGHV sequences that were shared by the BALB/c strain and any of the four wild-derived strains. The origins of IGHV genes in the C57BL/6, MSM/MsJ and BALB/c strains therefore remain unclear. These unexpected similarities and differences highlight our lack of understanding of the antibody gene loci of the laboratory mouse, with implications for the interpretation of strain-specific differences in models of antibody-mediated diseases, and of Adaptive Immune Receptor Repertoire sequencing (AIRR-seq) data. These results also suggest that a position-based immunoglobulin gene nomenclature may be unworkable in the mouse.

show abstract

Section: Introductionmentioning

confidence: 99%

A comparison of immunoglobulin IGHV, IGHD and IGHJ genes in wild-derived and classical inbred mouse strains

Watson

Kos

Gibson

et al. 2019

Preprint

Self Cite

View full text Add to dashboard Cite

show abstract

“…When such mismatches are repeatedly seen in a large set of VDJ rearrangements having diverse CDR3 regions and amplified from a single individual, the inference of a previously undiscovered gene polymorphism may be made with confidence. The discovery of allelic variants by inference is now a feature of many human repertoire studies, and this is facilitated by a number of recently developed utilities . When this approach was applied to the mouse, the outcome was quite unexpected.…”

Section: Introductionmentioning

confidence: 99%

“…The discovery of allelic variants by inference is now a feature of many human repertoire studies, and this is facilitated by a number of recently developed utilities. [22][23][24][25] When this approach was applied to the mouse, the outcome was quite unexpected.…”

Section: Introductionmentioning

confidence: 99%

A comparison of immunoglobulin IGHV, IGHD and IGHJ genes in wild‐derived and classical inbred mouse strains

et al. 2019

View full text Add to dashboard Cite

The genomes of classical inbred mouse strains include genes derived from all three major subspecies of the house mouse, Mus musculus. We recently posited that genetic diversity in the immunoglobulin heavy chain (IGH) gene loci of C57BL/6 and BALB/c mice reflects differences in subspecies origin. To investigate this hypothesis, we conducted high‐throughput sequencing of IGH gene rearrangements to document IGH variable (IGHV), joining (IGHJ) and diversity (IGHD) genes in four inbred wild‐derived mouse strains (CAST/EiJ, LEWES/EiJ, MSM/MsJ and PWD/PhJ) and a single disease model strain (NOD/ShiLtJ), collectively representing genetic backgrounds of several major mouse subspecies. A total of 341 germline IGHV sequences were inferred in the wild‐derived strains, including 247 not curated in the international ImMunoGeneTics information system. By contrast, 83/84 inferred NOD IGHV genes had previously been observed in C57BL/6 mice. Variability among the strains examined was observed for only a single IGHJ gene, involving a description of a novel allele. By contrast, unexpected variation was found in the IGHD gene loci, with four previously unreported IGHD gene sequences being documented. Very few IGHV sequences of C57BL/6 and BALB/c mice were shared with strains representing major subspecies, suggesting that their IGH loci may be complex mosaics of genes of disparate origins. This suggests a similar level of diversity is likely present in the IGH loci of other classical inbred strains. This must now be documented if we are to properly understand interstrain variation in models of antibody‐mediated disease.

show abstract

“…The V/D/J gene of a sequence with higher similarity to a novel allele than to the reference gene was reassigned to the novel allele. For each sample a genotype was constructed from sequences with a single assignment (only one best match), using TIgGER adapted for Bayesian approach (Gadala-Maria et al, 2018). Overall, 25 novel V alleles were identified and set as part of individuals' genotypes.…”

Section: Methodsmentioning

confidence: 99%

“…We exploited the fact that only naïve cells were sequenced to infer and study the characteristics of their germline IGH locus. After filtering out five samples with low coverage (< 2000 sequences), personal genotypes of the IGH regions were constructed using a Bayesian genotype approach (Gadala-Maria et al, 2018). To eliminate further potential biases genotype construction was based on unique sequences with at most three mutations in their V region and no mutations in the D region.…”

Section: Relative Gene Usage Can Indicate Gene Deletionsmentioning

confidence: 99%

Mosaic deletion patterns of the human antibody heavy chain gene locus as revealed by Bayesian haplotyping

Gidoni

Snir

Peres

et al. 2018

Preprint

Self Cite

View full text Add to dashboard Cite

Analysis of antibody repertoires by high throughput sequencing is of major importance in understanding adaptive immune responses. Our knowledge of variations in the genomic loci encoding antibody genes is incomplete, mostly due to technical difficulties in aligning short reads to these highly repetitive loci. The partial knowledge results in conflicting V-D-J gene assignments between different algorithms, and biased genotype and haplotype inference. Previous studies have shown that haplotypes can be inferred by taking advantage of IGHJ6 heterozygosity, which is observed in approximately one third of the population. Here, we propose a robust novel method for determining V-D-J haplotypes by adapting a Bayesian framework. In addition to IGHJ-based inference, our method extends haplotype inference also to IGHD-and IGHV -based, in heterozygous individuals. We can thus infer complex genetic events like deletions and copy number variations in a much larger fraction of the population. We tested our method on the largest data set, to date, of naïve B-cell repertoires. We present evidence for allele usage bias, as well as a mosaic deletion pattern of IGHD and IGHV genes across the population. The inferred haplotypes and deletion patterns may have clinical implications for genetic predispositions to diseases. Our findings greatly expand the knowledge that can be extracted from antibody repertoire sequencing data.

show abstract

Identification of subject-specific immunoglobulin alleles from expressed repertoire sequencing data

Cited by 22 publications

References 33 publications

A comparison of immunoglobulin IGHV, IGHD and IGHJ genes in wild-derived and classical inbred mouse strains

A comparison of immunoglobulin IGHV, IGHD and IGHJ genes in wild-derived and classical inbred mouse strains

A comparison of immunoglobulin IGHV, IGHD and IGHJ genes in wild‐derived and classical inbred mouse strains

Mosaic deletion patterns of the human antibody heavy chain gene locus as revealed by Bayesian haplotyping

Contact Info

Product

Resources

About