The use of non-functional clonotypes as a natural calibrator for quantitative bias correction in adaptive immune receptor repertoire profiling

Smirnova, Anastasia O; Miroshnichenkova, A; Olshanskaya, Yulia; Maschan, Michael; Lebedev, Yuri B.; Chudakov, Dmitriy M.; Mamedov, Ilgar Z.; Komkov, Alexander

doi:10.7554/elife.69157

Cited by 7 publications

(4 citation statements)

References 36 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…(ii) Combine LigO with sequencing read simulators (Huang et al 2012; Gourlé et al 2019) to incorporate PCR or sequencing errors. This type of error simulation is essential for studying artificially induced diversity versus biological clonal expansion and batch effects (Smirnova et al 2023; Pavlović et al 2022). (iii) Moreover, realistic simulation of cross-reactivity (i.e., the simulation of multiple immune signals per sequence) and chain pairing require further biological insights.…”

Section: Discussionmentioning

confidence: 99%

Simulation of adaptive immune receptors and repertoires with complex immune information to guide the development and benchmarking of AIRR machine learning

Chernigovskaya,

Pavlović,

Kanduri

et al. 2023

Preprint

View full text Add to dashboard Cite

Machine-learning methods (ML) have shown great potential in the adaptive immune receptor repertoire (AIRR) field. However, there is a lack of large-scale ground-truth experimental AIRR data suitable for AIRR-ML-based disease diagnostics and therapeutics discovery. Simulated ground-truth AIRR data are required to complement the development and benchmarking of robust and interpretable AIRR-ML approaches where experimental data is inaccessible or insufficient as of yet. The challenge for simulated data to be useful is the ability to incorporate key features observed in experimental repertoires. These features, such as complex antigen or disease-associated immune information, cause AIRR-ML problems to be challenging. Here, we introduce LIgO, a modular software suite, which simulates AIRR data for the development and benchmarking of AIRR-based machine learning. LIgO incorporates different types of immune information both on the receptor and the repertoire level and preserves native-like generation probability distribution. Additionally, LIgO assists users in determining the computational feasibility of their simulations. We show two examples where LIgO simulation supports the development and validation of AIRR-ML methods: (1) how individuals carrying out-of-distribution immune information impacts receptor-level prediction performance and (2) how immune information co-occurring in the same AIRs have an impact on the performance of conventional receptor-level encoding and repertoire-level classification approaches. The LIgO software guides the advancement and assessment of interpretable AIRR-ML methods.

show abstract

Section: Discussionmentioning

confidence: 99%

Simulation of adaptive immune receptors and repertoires with complex immune information to guide the development and benchmarking of AIRR machine learning

Chernigovskaya,

Pavlović,

Kanduri

et al. 2023

Preprint

View full text Add to dashboard Cite

show abstract

“…Obtained clonotype tables were converted into VDJtools format using “Convert” function with “-S mixcr” parameter. The level of quantitative bias was checked using iROAR software ( 33 ). The TRBJ2 containing clonotypes with detected D1 region were extracted from the analyzed dataset using “FilterBySegment” function and then were separated to functional and nonfunctional clonotype tables using the function “FilterNonFunctional”.…”

Section: Methodsmentioning

confidence: 99%

Novel bimodal TRBD1-TRBD2 rearrangements with dual or absent D-region contribute to TRB V-(D)-J combinatorial diversity

Smirnova,

Miroshnichenkova,

Belyaeva

et al. 2023

Front. Immunol.

View full text Add to dashboard Cite

T-cell receptor (TR) diversity of the variable domains is generated by recombination of both the alpha (TRA) and beta (TRB) chains. The textbook process of TRB chain production starts with TRBD and TRBJ gene rearrangement, followed by the rearrangement of a TRBV gene to the partially rearranged D-J gene. Unsuccessful V-D-J TRB rearrangements lead to apoptosis of the cell. Here, we performed deep sequencing of the poorly explored pool of partial TRBD1-TRBD2 rearrangements in T-cell genomic DNA. We reconstructed full repertoires of human partial TRBD1-TRBD2 rearrangements using novel sequencing and validated them by detecting V-D-J recombination-specific byproducts: excision circles containing the recombination signal (RS) joint 5’D2-RS – 3’D1-RS. Identified rearrangements were in compliance with the classical 12/23 rule, common for humans, rats, and mice and contained typical V-D-J recombination footprints. Interestingly, we detected a bimodal distribution of D-D junctions indicating two active recombination sites producing long and short D-D rearrangements. Long TRB D-D rearrangements with two D-regions are coding joints D1-D2 remaining classically on the chromosome. The short TRB D-D rearrangements with no D-region are signal joints, the coding joint D1-D2 being excised from the chromosome. They both contribute to the TRB V-(D)-J combinatorial diversity. Indeed, short D-D rearrangements may be followed by direct V-J2 recombination. Long D-D rearrangements may recombine further with J2 and V genes forming partial D1-D2-J2 and then complete V-D1-D2-J2 rearrangement. Productive TRB V-D1-D2-J2 chains are present and expressed in thousands of clones of human antigen-experienced memory T cells proving their capacity for antigen recognition and actual participation in the immune response.

show abstract

“…Obtained clonotype tables were converted into VDJtools format using "Convert" function with "-S mixcr" parameter. The level of quantitative bias was checked using iROAR software [27]. The TRBJ2 containing clonotypes with detected D1 segment were extracted from the analyzed dataset using "FilterBySegment" function and then were separated to functional and nonfunctional clonotype tables using the function "FilterNonFunctional".…”

Section: Vddj Rearrangements Detectionmentioning

confidence: 99%

Non-canonical D1-D2 recombination produces two types of rearrangements and represents a conservative hidden stage of T-cell receptor beta chain generation

Smirnova

Miroshnichenkova

Belyaeva

et al. 2023

Preprint

View full text Add to dashboard Cite

T-cell receptor (TCR) diversity is generated by VDJ recombination. The classical course of TCR beta (TRB) chain production starts with D and J segment recombination and finishes with subsequent recombination between the resulting DJ junction and V segment. In this study, we performed deep sequencing of poorly explored incomplete TRBD1 to TRBD2 rearrangements in T-cell genomic DNA. Here we reconstructed full human TRB DD rearrangements repertoires for the first time. We validated its authenticity by detecting excision circles with RSS (recombination signal sequence) junctions. The found rearrangements generated in compliance with the classical 12/23 rule are common for humans, rats, and mice, and contain typical VDJ footprints: random nucleotide deletions and insertions. Detected bimodal distribution of DD junctions indicates two active recombination sites producing long and short rearrangements. Unlike long DD rearrangements, the short ones have unusual origin resulting from non-canonical intrachromosomal RSSs junctions formation. Identified DD rearrangement leads to the deletion of J1 and C1 segments and creates a diverse hybrid D segment that can further recombine with J2 and V segments. However, in most complete TRB genes, the traces of DD junctions are blurred by cutting in DDJ and VDDJ rearrangements which made this stage of VDJ recombination mostly hidden in the final completely rearranged TRB gene.

show abstract

The use of non-functional clonotypes as a natural calibrator for quantitative bias correction in adaptive immune receptor repertoire profiling

Cited by 7 publications

References 36 publications

Simulation of adaptive immune receptors and repertoires with complex immune information to guide the development and benchmarking of AIRR machine learning

Simulation of adaptive immune receptors and repertoires with complex immune information to guide the development and benchmarking of AIRR machine learning

Novel bimodal TRBD1-TRBD2 rearrangements with dual or absent D-region contribute to TRB V-(D)-J combinatorial diversity

Non-canonical D1-D2 recombination produces two types of rearrangements and represents a conservative hidden stage of T-cell receptor beta chain generation

Contact Info

Product

Resources

About