An Open-Source Toolkit That Powers the Genome-Wide Analysis of Mature B-Cell Lymphomas (GAMBL) Project

Abstract Introduction: Genome- and transcriptome-wide analyses continue to enhance our understanding of the molecular pathogenesis cancer. In lymphomas, this has enabled identification hundreds recurrently mutated genes, highlighting genetic heterogeneity relationships both within among clinical entities. While growing availability lymphoma genomic data sets can be leveraged integrate into diagnostic testing trials, ability rapidly process in a reproducible manner serves as barrier goal. To end, we developed suite tools Lymphoid Cancer Research modules (LCR-modules) facilitate discovery novel drivers features cancers perform quantitative comparisons between disease We demonstrate here how toolkit meta-analysis involving genome-wide profiles 3330 patients. Methods: assembled collection whole genome, exome, RNA sequencing from combination controlled-access repositories ongoing projects at BC Cancer. The scope analysis mature B-cell lymphomas (GAMBL) project includes cell lines patient tumors all common B neoplasms, comprising total 4612 samples project, open-source custom bioinformatics (https://github.com/LCR-BCCRC/lcr-modules) that leverages Snakemake workflow management system lymphoma-centric for annotation mutation types, receptor repertoires aSHM targets relevant non-coding mutations, RNA-seq with batch correction normalization. Individual are configured create an automated, scalable, runs each step dictated by new data. cohort-level integrative across entities handled R package GAMBLR, which facilitates open-ended visualizations. Results: Simple somatic mutations (SSM) were detected using utilizes four algorithms identify high-confidence variants validated default thresholds filtering germline FFPE-associated artifacts, allowing processing without matched normal tissue. This automated facilitated genes significantly broadened aberrant hypermutation (aSHM) other mutations. Specifically, HNRNPU, STAT3, TFAP4, RRAGC found relatively low frequencies, their presence is distinct feature certain or subgroups types (Figure 1A). hypermutated regions tool Rainstorm. As result, able detect sites preferentially single entity, such transcription start site BACH2, lower rates than target but more BL compared 1B). Combining discovered allowed us explore establish Burkitt follicular lymphoma. SV conducted Manta, GRIDSS, JaBbA downstream GAMBLR. most SVs identified targeting MYC, BCL2, CCND1. Unsurprisingly, translocation partner was immunoglobulin heavy chain, BCL6-FOXP1, CD274-BACH2, BCL6-RHOH translocations DLBCLs MYC-BCL6 BLs identified, others 1C). Conclusions: present modularized scalable transcriptomic it successfully deployed thousands tumour known biology. represents important advancement reproducibility will translation discoveries. Figure 1 1. Disclosures Grande: Sage Bionetworks: Current Employment. Coyle: Allakos, Inc.: Consultancy. Steidl: AbbVie: Consultancy; Trillium Therapeutics: Funding; Epizyme: Seattle Genetics: Curis Bayer: Bristol-Myers Squibb: Funding. Scott: Abbvie: NanoString Technologies: Patents & Royalties: Patent describing measuring proliferation signature MCL gene expression profiling.; Celgene: AstraZeneca: Incyte: Janssen: Consultancy, Rich/Genentech: Cancer: assigning DLBCL COO profiling--licensed Technologies. profiling. . Morin: Royalties; Foundation Lymphoma Research: Membership on entity's Board Directors advisory committees.