Fine mapping of V(D)J recombinase mediated rearrangements in human lymphoid malignancies
© Halper-Stromberg et al.; licensee BioMed Central Ltd. 2013
Received: 9 January 2013
Accepted: 6 August 2013
Published: 19 August 2013
Lymphocytes achieve diversity in antigen recognition in part by rearranging genomic DNA at loci encoding antibodies and cell surface receptors. The process, termed V(D)J recombination, juxtaposes modular coding sequences for antigen binding. Erroneous recombination events causing chromosomal translocations are recognized causes of lymphoid malignancies. Here we show a hybridization based method for sequence enrichment can be used to efficiently and selectively capture genomic DNA adjacent to V(D)J recombination breakpoints for massively parallel sequencing. The approach obviates the need for PCR amplification of recombined sequences.
Using tailored informatics analyses to resolve alignment and assembly issues in these repetitive regions, we were able to detect numerous recombination events across a panel of cancer cell lines and primary lymphoid tumors, and an EBV transformed lymphoblast line. With reassembly, breakpoints could be defined to single base pair resolution. The observed events consist of canonical V(D)J or V-J rearrangements, non-canonical rearrangements, and putatively oncogenic reciprocal chromosome translocations. We validated non-canonical and chromosome translocation junctions by PCR and Sanger sequencing. The translocations involved the MYC and BCL-2 loci, and activation of these was consistent with histopathologic features of the respective B-cell tumors. We also show an impressive prevalence of novel erroneous V-V recombination events at sites not incorporated with other downstream coding segments.
Our results demonstrate the ability of next generation sequencing to describe human V(D)J recombinase activity and provide a scalable means to chronicle off-target, unexpressed, and non-amplifiable recombinations occurring in the development of lymphoid cancers.
KeywordsV(D)J recombination Oncogenic translocation Lymphoid tumors MYC BCL-2 V replacement
In humans, the number of unique immunoglobulin (Ig) and T-cell receptors (TCRs) approaches 1012. Much of this diversity is generated through the combinatorial assembly of antigen receptor genes from discrete DNA segments by the process of V(D)J recombination. Aberrant V(D)J recombination is one mechanism responsible for chromosomal translocations associated with lymphoid malignancies. Characterization of canonical and aberrant V(D)J rearrangements is fundamental to understanding the primary antigen receptor repertoire and the pathogenesis of lymphoid malignancies. Reflecting this interest and developments in sequencing technology, the International Immunogenetics Information System (IMGT) blast database has increased in size from ~21 Mb to ~63 Mb of human sequence data since 2003 . Even at this size, however, the database comprises only 746 functionally recombined human alleles and 321 non-functional alleles or pseudogenes.
Identification of V(D)J recombination products by next generation sequencing techniques is an attractive way to obtain a more comprehensive sample of the estimated repertoire . However, sequence properties that define Ig and TCR loci also make their interpretation from short read technology challenging. Standard practices like removal of poorly aligned reads must be modified to accommodate the specific structural changes expected during V(D)J recombination and the prevalence of regional segmental duplications. Complexities associated with this task are well recognized .
We have developed a targeted method to selectively sequence V(D)J recombination events. It has the advantage of revealing off-target and non-canonical rearrangements that may be unexpressed or unamplifiable with primer pairs designed to detect canonical rearrangements. We demonstrate the method in a panel of primary lymphoid tumors and lymphoid cell lines and report high-resolution sequences comprising canonical and unknown, non-canonical rearrangements as well as two chromosome translocations mediated by erroneous V(D)J recombination events. We expect that this approach will provide deeper insight into the role V(D)J recombination plays in pathogenesis of lymphoproliferative disease.
Ambiguous alignments (i.e., reads matching multiple genomic positions) were frequently encountered. This reflects inherent properties of antigen receptor loci, which contain arrays of homologous gene segments at several genomic loci. Baits and surrounding contexts were enriched for segmental duplications about 6-fold above the genome-wide level. Where possible we resolved ambiguous alignments with sensitive realignment. We frequently could validate events within segmental duplications, provided the two junctional sequences were not homologous. Of the 8 non-canonical events reported here, 4 were within segmental duplications, and of the 18 canonical events, 9 were within segmental duplications. A more detailed description of how we distinguished true structural variants from alignment artifacts over these intervals and an associated R package to aid in visualizing alternate alignments will be described in a companion paper (Halper-Stromberg, et al. In preparation).
Canonical V(D)J and V-J events
Observed V(D)J recombination events
A) pre-B ALL
B) chronic T-Cell leukemia
The immunoglobulin kappa light chain locus (IGK) and the locus encoding T-cell receptor β chain (TRB) loci followed the same pattern, with recurrent utilization of J segments but not of the more numerous V segments (IGK) or V or D segments (TRB). The IGK locus J1 and J2 RSSs were recombined one time each in the LCL, out of 2 observed V-J joints, and once and 2 times in neoplastic B-cells, respectively, out of 3 observed V-J joints. Use of TRB J2 was observed in two of three T-cell neoplasms studied. We observed two events using the same variable segment (V3-19) at the immunoglobulin lambda light chain locus (IGL); these were recombined with different J elements.
Most (6/9) canonical, non-IGH rearrangements detected were inversions. We attribute this to a bias of our method towards inversion detection. Capturing inversional rearrangements, which retain RSS sequences, is an expected outcome as probes were designed to hybridize RSS sequences directly. Deletional rearrangement detection, conversely, requires sequencing rearranged segments adjacent to a captured RSS not participating in the rearrangement.
Four of the inversion junctions observed represented signal joints and two represented coding joints. One coding joint resulted in a productive V-J allele . The other coding joint resulted in a productive V(D)J allele . Four of the 6 inversions occurred in a single sample, 3 in neoplastic B-cells, and 1 in a T cell neoplasm. The LCL sample exhibited 2 inversions, both at IGK, one of these being the largest intra-chromosomal rearrangement observed (Figure 2). This larger inversion involved a >1 Mb segment of 2p11.2 bringing IGKJ1 and IGKV1D-43 together.
All five canonical rearrangements in the IGK locus were inversions, including one in a Burkitt-like lymphoma using a J element < 1 Kb centromeric of the t(2;8) breakpoint seen on a different allele of the same sample (Additional file 1: Figure S1). This event involved a 740 Kb segment of 2p11.2, bringing IGKJ2 together with IGKV1D-39. The T-cell inversion in Loucy cells was the only inversional joining of V, D, and J segments that we observed (Additional file 1: Figure S2). This rearrangement we infer occurred in 2 steps: a deletion bringing TRBJ2-1 into contact with an upstream D element in 7q34, and the subsequent joining of the DJ unit to TRBV5-6 via inversion.
Non-Canonical and lineage inappropriate rearrangements
Of the other two non-canonical events, one was a lineage-inappropriate rearrangement, a V to D rearrangement via deletion at IGH in Loucy cells, which are derived from a precursor T-cell acute lymphoblastic leukemia . The other non-canonical recombination was a putative V-J rearrangement at TRB in a chronic T-cell leukemia sample, apparently lacking the expected D element between the V and J (Additional file 1: Figure S6). This event involved a 16 Kb inversion in 7q34 rearranging TRBJ2-1 with TRBV30. We obtained sequence from the V RSS to J RSS junction but not the V-J segment junction, inferring that neither of the two TRBD elements had recombined with J elements from the TRBJ2 cluster preceding the inversion, as this would have deleted the 12 bp spacer RSS flanking TRBJ2-1 that we observed. The recombination did not violate the 12–23 recombination rule, but a D-J rearrangement should have preceded incorporation of the V segment .
We observed addition of N-region (non-coded) nucleotides at all seven validated junctions involving at least one coding segment. These included 4 short additions between 2 and 5 bp across 3 samples, 2 in LCL, 1 in DB cells, and 1 in Loucy cells; 2 intermediate size additions of 8 and 11 bp both in DB cells; and one long addition of 23 bp in ARH-77. The activity of terminal deoxynucleotidyl transferase (TdT) enzyme can explain these additions, although the addition in ARH-77 is unusually long. N regions were highly specific to coding segments and not seen at any of the 6 RSS-RSS junctions in which we obtained split-reads, including the five IGK canonical inversions and the V to J rearrangement at TRB in the chronic T-cell leukemia sample.
Oncogenic interchromosomal translocations
In two samples, large numbers of reads mapped to loci on other chromosomes not targeted for sequencing, reflecting possible V(D)J recombinase mediated translocations with breakpoints near RSS. The first off-target sequences mapped on the telomeric side of the MYC oncogene on chromosome 8 and occurred with partial read or read pairs mapping on the centromeric side of IGKJ4 on chromosome 2. This was consistent with a t(2;8) translocation. This lesion was observed in a pediatric patient who had been immunosuppressed after receiving an organ transplant, and is consistent with having contributed to the development of a Burkitt-like lymphoma in this person. Genomic DNA from fresh frozen primary cells was used for our analysis.
Previously described IGH-BCL-2 translocation breakpoints in B-cell neoplasms are similar to ours but not exactly matching. We searched 2 databases to determine the novelty of our breakpoints; breaks within IGHJ5 are common and closely match ours, whereas the other breakpoints involved are less often seen and do not match as closely. In dbCRID: Database of Chromosomal Rearrangements in Disease, we found sequence at the breakpoints on der(14) for 6 diffuse large B-cell lymphoma patient . On the centromeric side, these clustered in a 22 bp window within IGHJ6, approximately 600 bp centromeric to our breakpoint within IGHJ5 . On the telomeric side, breakpoints clustered in a 50 bp window within the 3′ UTR of BCL-2, approximately 27.5 Kb telomeric of our breakpoint.
IGH-BCL-2 translocation breakpoints closer to the ones we observed were found in a chromosomal rearrangement breakpoint database containing 551 t(14;18) entries . We found 2 exact matches to our der(14) breakpoint on the centromeric side (within the IGHJ5 coding segment), both in non-malignant B cell samples. Each had sequence additions at the breakpoints albeit different from ours. Of follicular and diffuse large B cell lymphoma entries in the database, 163 had a breakpoint within 1Kb of our der(14) IGHJ5 breakpoint, with 34 matching within 5 bp. We did not find exact, or as nearly exact, breakpoint matches in the database for other t(14;18) junctional sequences. For the der(18) IGHD5-12 breakpoint the nearest entry was 9 bp away for a non-malignant B cell sample and 1840 bp away for a B cell lymphoma. For our breakpoints centromeric of BCL-2, the nearest entry, from a B cell lymphoma, was 683 and 695 bp away from our der(14) and der(18) breakpoints, respectively. In a pattern similar to dbCRID, 90% of BCL-2 breakpoints in the database were in the 3′ UTR, the vast majority clustering in a 150 bp window. Only 3% of BCL-2 breakpoints were further downstream of the gene than ours. Nonetheless our der(14) conformed in structure to the predominant der(14) model from the database, whereby the IGHJ locus adjoins BCL-2, coding sequence intact.
Because the IGH-BCL-2 translocation had not to our knowledge been reported previously in this cell line, we performed standard metaphase karyotyping analysis. Seventeen metaphases were evaluated in this near tetraploid cell line. The cells showed the following composite, complex karyotype: 77 ~ 80 < 4n>, XXYY, -1, -2, -3, -3, -4, add(4)(p15.2), -5, add(5)(p13), del(5)(q22q33), -6,+7, der(8;?19)(p10;?q10), -9, -9, -10, add(10)(q22), -12, -13, -13, add(13)(q34), -14, t(14;18)(q32;q21), -15, -16, der(18)t(14;18)(q32;q21), +20, add(22)(q11.2), +mar1, +mar2[cp17]. The abnormalities observed in the karyotype aside from t(14;18) were not seen in the sequencing data as we had no baits covering these loci.
A relatively large portion of events was from the LCL sample. This is consistent with the oligoclonal or heterogeneous derivation of this cell line, which provides more opportunity to detect events than clonal, neoplastic cells. The finding suggests that sequence capture methods may be applied to mixed lymphoid populations, including oligoclonal proliferations in the context of disease or polyclonal populations at various stages of lymphoid development. Understanding detection limits in populations of cells and applying the method to single cells are areas of future investigation.
For all junctional reads from coding segments, we observed N-region addition of nucleotides at the breakpoints, which we attribute to terminal deoxynucleotidyl transferase (TdT) enzyme activity (Figures 3 and 5 B, Additional file 1: Figures S2, S4 and S5). We did not see this same addition for junctions overlapping only RSS elements, with RSS-RSS junctions mapping exactly to the reference genome with no intervening, untemplated bases (Figures 2 and 4, Additional file 1: Figures S1 and S6). Most nucleotide additions adjacent to a coding segment were between 2 and 5 bases, although we observed 3 longer additions. For one of these exceptions, a recombination by interstitial deletion, we observed 23 N-region nucleotides that did not align to either side of the juxtaposed sequences (Additional file 1: Figure S5). Although this is an unusual length of nucleotides for TdT addition, it is unclear if dysregulated TdT activity is otherwise related to the proliferation.
We expect that inspection of V(D)J recombination using high throughput sequencing will provide a more complete picture of rearrangement at antigen receptor loci. Perhaps more importantly, next generation capture sequencing approaches such as ours may prove powerful in the study of how lymphoid populations change in response to antigenic selection or during clonal evolution of malignancies .
Our experience underscores the value of targeted methods, both in sequencing and in analytical approaches. First, we have optimized cost and output with a sequence-based targeted pull down method, enabling deep sequencing of selected regions while minimizing data generation and processing time. Excellent coverage of desired sequences with effective exclusion of the remainder of the genome demonstrates the exquisite specificity of RSS sequences for genomic DNA capture. Secondly, in analytical aspects, we have balanced throughput with accuracy and shown the utility of contextual inspections of ambiguous aligning sequences.
Other methods of sequencing products of V(D)J recombination, even those leveraging considerable sequencing throughput, frequently focus on expressed products present in RNA/cDNA fractions or products that are amplifiable from genomic DNA sequences using primers anticipating canonical rearrangements of coding segments. Our method has the advantage of independence from these. As such, it should prove especially useful for finding off target and non-canonical rearrangements, such as the V(D)J recombinase-mediated chromosomal translocations and non-canonical V-V deletion events we described here. The latter example, which occurred with unexpected frequency in our small sample set, reflects a category of events that have not been well described outside of rearranged V-replacement. Finally, this advantage of our approach may facilitate the detection of V(D)J recombinase-mediated translocations in preneoplastic leukemia samples  or the identification of transposition events [20, 21] in lymphocyte development and in lymphoid pathologies.
SureSelect library design
Recombination signal sequences (RSSs) recognized by RAG recombinase were downloaded from the international ImMunoGeneTics (IMGT) database  and used to recover corresponding genomic coordinates by alignment to the human reference assembly (hg19/GRCh37, February 2009) with Blat . For each matching sequence, a 200 bp target window immediately flanking V(D)J coding segments was defined, extending across the entire RSS and into non-coding sequence 3′ of the RSS; these were merged to 440 genomic intervals for five-fold Agilent SureSelect bait tiling (Agilent Technologies, Santa Clara CA). A total of 2461 120-mer baits were designed to pull-down these sequences using eArray (Additional file 4).
The Johns Hopkins Medicine Institutional Review Board granted approval for the study with reference number NA_00050660. Kathleen H. Burns, a clinical hematopathologist and co-author of this work, acquired cells from the clinical flow cytometry lab at Johns Hopkins Hospital. Samples were collected for diagnostic purposes with research material appropriated from the excess of what was needed for diagnostics. No additional proce-dures were performed to collect these samples, and so there was no associated medical risk to patients. Aliquots used for this study would have otherwise been discarded from the clinical flow cytometry lab. Samples were de-identified so that no patient information would follow samples to the lab. The Johns Hopkins Institutional Review Board approved their use without patient consent.
We identified rearrangements in four primary lymphoid malignancies; four cancer cell lines (American Type Culture Collection, Manassas, VA); and one EBV-transformed lymphoblastoid cell line (HuRef, Coriell Institute, Camden, NJ). No rearrangements were identified in two primary lymphoid malignancies studied, a chronic lymphocytic leukemia and a precursor T-cell acute lymphoblastic leukemia. The four primary neoplasias included a precursor B-cell acute lymphoblastic leukemia; a post transplant lymphoproliferative disorder consistent with a high grade, Burkitt-like lymphoma; a mantle cell lymphoma; and a chronic lymphocytic leukemia of the T-cell lineage. The four cell lines included the Ramos Burkitt lymphoma line; the DB diffuse large B-cell lymphoma of follicle center cell phenotype line; the ARH-77 plasma cell leukemia line; and Loucy acute precursor T-cell lymphoblastic lymphoma line.
Targeted DNA library preparation and sequencing
High molecular weight gDNA was obtained from viable cells or fresh frozen primary cells by phenol-chloroform extraction and ethanol precipitation. For each sample, we sonicated gDNA to 500 bp (median size). Sample DNA was end-repaired using NEBNext End Repair Module (New England Biolabs, Ipswich MA), purified, dA-tailed, and ligated to index-specific, paired-end adapters from Illumina’s Multiplexing Oligonucleotide Kit (Illumina Inc, San Diego CA). Agilent’s SureSelect Target Enrichment System Kit for Illumina Paired-End Multiplexed Sequencing was used to complete library preparation and pull-down. Adapter-ligated DNA samples were PCR-amplified, purified, and hybridized to custom-designed RNA baits to capture our targeted DNA. Captured DNA was pulled down using Dynal MyOne Streptavidin T1 magnetic beads (Life Technologies, Carlsbad CA). Pulled-down DNA samples were index tagged using Illumina’s Multiplexing Oligonucleotide Kit. Products were purified with AMPure XP magnetic beads to produce targeted libraries. HudsonAlpha Institute for Biotechnology (Huntsville, AL) performed quality control, pooling, and sequencing of our indexed samples. The multiplexed library was sequenced in one lane of an Illumina HiSeq generating 7,339,278 100 bp paired-end reads of median insert size 254 and median coverage 173X across the 440 bait regions.
We aligned reads to hg19, used HYDRA to identify candidate recombinations , and visualized sites using the Integrative Genomics Viewer . Breakpoints were determined by splitting reads with at least one member of a pair mapping to one of the two loci for an event, and remapping. We visualized split-reads using Jalview .
We designed primers to amplify across rearrangement breakpoints, calling an event validated if we observed the expected band and no similarly sized band within control DNA. We validated all reported non-canonical and some canonical events (Additional files 2, 3, 5).
Histopathology and flow cytometry
Morphologic features of primary tumors were studied by hematoxylin and eosin staining of tissue sections. Phenotyping by immunohistochemistry or flow cytometry was performed in the clinical laboratories at the Johns Hopkins Hospital as part of the routine diagnostic work-up. Lesions were classified according to the 2008 World Health Organization diagnostic criteria.
Recombination signal sequence
T-cell receptor β chain
Immunoglobulin kappa light chain
Immunoglobulin lambda light chain
Immunoglobulin heavy chain
Lymphoblastoid cell line
terminal deoxynucleotidyl transferase.
We thank Dr. Rafael Irizarry for helpful discussions regarding strategies for alignment and evaluation of results.
This work was supported by a Career Award for Medical Scientists from the Burroughs Wellcome Foundation (to K.H.B.) and The National Institutes of Health (R01HG005220).
- Alamyar E, Duroux P, Lefranc MP, Giudicelli V: IMGT((R)) tools for the nucleotide analysis of immunoglobulin (IG) and T cell receptor (TR) V-(D)-J repertoires, polymorphisms, and IG mutations: IMGT/V-QUEST and IMGT/HighV-QUEST for NGS. Methods Mol Biol. 2012, 882: 569-604. 10.1007/978-1-61779-842-9_32.View ArticlePubMedGoogle Scholar
- Lefranc MP, Giudicelli V, Busin C, Bodmer J, Muller W, Bontrop R, Lemaitre M, Malik A, Chaume D: IMGT, the International ImMunoGeneTics database. Nucleic Acids Res. 1998, 26 (1): 297-303. 10.1093/nar/26.1.297.PubMed CentralView ArticlePubMedGoogle Scholar
- Fischer N: Sequencing antibody repertoires: the next generation. mAbs. 2011, 3 (1): 17-20. 10.4161/mabs.3.1.14169.PubMed CentralView ArticlePubMedGoogle Scholar
- Watson CT, Breden F: The immunoglobulin heavy chain locus: genetic variation, missing data, and implications for human disease. Genes Immun. 2012, 13 (5): 363-373. 10.1038/gene.2012.12.View ArticlePubMedGoogle Scholar
- Nilsson K, Klein G: Phenotypic and cytogenetic characteristics of human B-lymphoid cell lines and their relevance for the etiology of Burkitt’s lymphoma. Adv Cancer Res. 1982, 37: 319-380.View ArticlePubMedGoogle Scholar
- Wasserman R, Ito Y, Galili N, Yamada M, Reichard BA, Shane S, Lange B, Rovera G: The pattern of joining (JH) gene usage in the human IgH chain is established predominantly at the B precursor cell stage. J Immunol. 1992, 149 (2): 511-516.PubMedGoogle Scholar
- Huber C, Klobeck HG, Zachau HG: Ongoing V kappa-J kappa recombination after formation of a productive V kappa-J kappa coding joint. Eur J Immunol. 1992, 22 (6): 1561-1565. 10.1002/eji.1830220632.View ArticlePubMedGoogle Scholar
- Malissen M, McCoy C, Blanc D, Trucy J, Devaux C, Schmitt-Verhulst AM, Fitch F, Hood L, Malissen B: Direct evidence for chromosomal inversion during T-cell receptor beta-gene rearrangements. Nature. 1986, 319 (6048): 28-33. 10.1038/319028a0.View ArticlePubMedGoogle Scholar
- Fanning L, Bertrand FE, Steinberg C, Wu GE: Molecular mechanisms involved in receptor editing at the Ig heavy chain locus. Int Immunol. 1998, 10 (2): 241-246. 10.1093/intimm/10.2.241.View ArticlePubMedGoogle Scholar
- Szczepanski T, Pongers-Willemse MJ, Langerak AW, Harts WA, Wijkhuijs AJ, Van Wering ER, Van Dongen JJ: Ig heavy chain gene rearrangements in T-cell acute lymphoblastic leukemia exhibit predominant DH6-19 and DH7-27 gene usage, can result in complete V-D-J rearrangements, and are rare in T-cell receptor alpha beta lineage. Blood. 1999, 93 (12): 4079-4085.PubMedGoogle Scholar
- Born W, Yague J, Palmer E, Kappler J, Marrack P: Rearrangement of T-cell receptor beta-chain genes during T-cell development. Proc Natl Acad Sci USA. 1985, 82 (9): 2925-2929. 10.1073/pnas.82.9.2925.PubMed CentralView ArticlePubMedGoogle Scholar
- Kroenlein H, Schwartz S, Reinhardt R, Rieder H, Molkentin M, Gokbuget N, Hoelzer D, Thiel E, Burmeister T: Molecular analysis of the t(2;8)/MYC-IGK translocation in high-grade lymphoma/leukemia by long-distance inverse PCR. Genes Chromosome Canc. 2012, 51 (3): 290-299. 10.1002/gcc.21915.View ArticleGoogle Scholar
- Lipford E, Wright JJ, Urba W, Whang-Peng J, Kirsch IR, Raffeld M, Cossman J, Longo DL, Bakhshi A, Korsmeyer SJ: Refinement of lymphoma cytogenetics by the chromosome 18q21 major breakpoint region. Blood. 1987, 70 (6): 1816-1823.PubMedGoogle Scholar
- Bakhshi A, Wright JJ, Graninger W, Seto M, Owens J, Cossman J, Jensen JP, Goldman P, Korsmeyer SJ: Mechanism of the t(14;18) chromosomal translocation: structural analysis of both derivative 14 and 18 reciprocal partners. Proc Natl Acad Sci U S A. 1987, 84 (8): 2396-2400. 10.1073/pnas.84.8.2396.PubMed CentralView ArticlePubMedGoogle Scholar
- Kong F, Zhu J, Wu J, Peng J, Wang Y, Wang Q, Fu S, Yuan LL, Li T: dbCRID: a database of chromosomal rearrangements in human diseases. Nucleic Acids Res. 2011, 39 (Database issue): D895-D900.PubMed CentralView ArticlePubMedGoogle Scholar
- Matolcsy A, Casali P, Warnke RA, Knowles DM: Morphologic transformation of follicular lymphoma is associated with somatic mutation of the translocated Bcl-2 gene. Blood. 1996, 88 (10): 3937-3944.PubMedGoogle Scholar
- Tsai AG, Lu H, Raghavan SC, Muschen M, Hsieh CL, Lieber MR: Human chromosomal translocations at CpG sites and a theoretical basis for their lineage and stage specificity. Cell. 2008, 135 (6): 1130-1142. 10.1016/j.cell.2008.10.035.PubMed CentralView ArticlePubMedGoogle Scholar
- Robison K: Application of second-generation sequencing to cancer genomics. Brief Bioinform. 2010, 11 (5): 524-534. 10.1093/bib/bbq013.View ArticlePubMedGoogle Scholar
- Wiemels JL, Cazzaniga G, Daniotti M, Eden OB, Addison GM, Masera G, Saha V, Biondi A, Greaves MF: Prenatal origin of acute lymphoblastic leukaemia in children. Lancet. 1999, 354 (9189): 1499-1503. 10.1016/S0140-6736(99)09403-9.View ArticlePubMedGoogle Scholar
- Agrawal A, Eastman QM, Schatz DG: Transposition mediated by RAG1 and RAG2 and its implications for the evolution of the immune system. Nature. 1998, 394 (6695): 744-751. 10.1038/29457.View ArticlePubMedGoogle Scholar
- Hiom K, Melek M, Gellert M: DNA transposition by the RAG1 and RAG2 proteins: a possible source of oncogenic translocations. Cell. 1998, 94 (4): 463-470. 10.1016/S0092-8674(00)81587-1.View ArticlePubMedGoogle Scholar
- Giudicelli V, Duroux P, Ginestoux C, Folch G, Jabado-Michaloud J, Chaume D, Lefranc MP: IMGT/LIGM-DB, the IMGT comprehensive database of immunoglobulin and T cell receptor nucleotide sequences. Nucleic Acids Res. 2006, 34 (Database issue): D781-D784.PubMed CentralView ArticlePubMedGoogle Scholar
- Kent WJ: BLAT–the BLAST-like alignment tool. Genome Res. 2002, 12 (4): 656-664.PubMed CentralView ArticlePubMedGoogle Scholar
- Quinlan AR, Clark RA, Sokolova S, Leibowitz ML, Zhang Y, Hurles ME, Mell JC, Hall IM: Genome-wide mapping and assembly of structural variant breakpoints in the mouse genome. Genome Res. 2010, 20 (5): 623-635. 10.1101/gr.102970.109.PubMed CentralView ArticlePubMedGoogle Scholar
- Robinson JT, Thorvaldsdottir H, Winckler W, Guttman M, Lander ES, Getz G, Mesirov JP: Integrative genomics viewer. Nat Biotechnol. 2011, 29 (1): 24-26. 10.1038/nbt.1754.PubMed CentralView ArticlePubMedGoogle Scholar
- Waterhouse AM, Procter JB, Martin DM, Clamp M, Barton GJ: Jalview Version 2–a multiple sequence alignment editor and analysis workbench. Bioinformatics. 2009, 25 (9): 1189-1191. 10.1093/bioinformatics/btp033.PubMed CentralView ArticlePubMedGoogle Scholar
This article is published under license to BioMed Central Ltd. This is an Open Access article distributed under the terms of the Creative Commons Attribution License (http://creativecommons.org/licenses/by/2.0), which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly cited.