Search

Burkitt lymphoma (BL) accounts for the majority of pediatric non-Hodgkin lymphomas (NHL) and is relatively rare but significantly more lethal when diagnosed in adults. The global incidence is highest in Sub-Saharan Africa, where Epstein-Barr virus (EBV) positivity is observed in 95% of all tumors. Both pediatric (pBL) and adult (aBL) cases are known to share some driver mutations, for example MYC translocations, which are seen in > 90% of cases. Sequencing efforts have identified many common somatic alterations that cooperate with MYC in lymphomagenesis with approximately 30 significantly mutated genes (SMG) reported thus far. Recent analyses revealed non-coding mutation patterns in pBL that were attributed to aberrant somatic hypermutation (aSHM). We sought to identify genomic and molecular features that may explain clinical disparities within and between aBL and pBL in an effort to delineate BL subtypes that may allow for the stratification of patients with shared pathobiology. Through comprehensive sequencing of BL genomes, we found additional SMGs, including more genetic features that associate with tumor EBV status, and established three new genetic subgroups that span pBL and aBL. Direct comparisons between pBL and aBL revealed only marginal differences and the mutational profiles were consistently better explained by EBV status. Using an unsupervised clustering approach to identify subgroupings within BL and diffuse large B-cell lymphoma (DLBCL), we have defined three genetic subgroups that predominantly comprise BL tumors. Akin to the recently defined DLBCL subgroups, each BL subgroup is characterized by combinations of common driver mutations and non-coding mutations caused by aSHM. Two of these subgroups and their prototypical genetic features (ID3 and TP53) had significant associations with patient outcomes that were different among the aBL and pBL cohorts. These findings highlight not only a shared pathogenesis between aBL and pBL, but also establish genetic subtypes within BL that serve to delineate tumors with distinct molecular features, providing a new framework for epidemiological studies, and diagnostic and therapeutic strategies.

Lifelong production of the many types of mature blood cells from less differentiated progenitors is a hierarchically ordered process that spans multiple cell divisions. The nature and timing of the molecular events required to integrate the environmental signals, transcription factor activity, epigenetic modifications, and changes in gene expression involved are thus complex and still poorly understood. We now show that the more primitive types of human cells in this system display a unique repressive H3K27me3 signature that is retained by mature lymphoid cells but is lost in terminally differentiated monocytes and erythroblasts. Additional intervention data implicate that control of this chromatin state change is a requisite part of the process, whereby normal human hematopoietic progenitor cells make lymphoid and myeloid fate decisions.

Computational biology requires the reading and comprehension of biological data files. Plain-text formats such as SAM, VCF, GTF, PDB and FASTA, often contain critical information that is obfuscated by the complexity of the data structures. bioSyntax (http://bioSyntax.org) is a freely available suite of syntax highlighting packages for vim, gedit, Sublime, and less, which aids computational scientists to parse and work with their data more efficiently.