Omic-14. Openpbta: An Open Pediatric Brain Tumor Atlas

Abstract

Pediatric brain tumors comprise a heterogeneous molecular and histological landscape that challenges most current precision-medicine approaches. While recent large-scale efforts to molecularly characterize distinct histological entities have dramatically advanced the field’s capacity to classify and further define molecular subtypes, developing therapeutic and less toxic molecularly-defined clinical approaches remains a challenge. To define new approaches to meet these challenges and advance scalable, shared biospecimen- and data-resources for pediatric brain tumors, the Children’s Brain Tumor Network and Pacific Pediatric Neuro-Oncology Consortium, in partnership with the Alex’s Lemonade Stand Foundation Childhood Cancer Data Lab, launched OpenPBTA, a global open science Pediatric Brain Tumor Atlas initiative to comprehensively define the molecular landscape of pediatric brain tumors. The initiative contains multi-modal analyses of research- and clinical-trial based DNA and RNA sequences from nearly 1,000 subjects (with 1,256 tumors) along with their longitudinal clinical data. The OpenPBTA’s open science framework for analysis tests the capacity of crowd-sourced collaborative architectures to advance more rapid, iterative and integrated discovery of the underlying mechanisms of disease across pediatric brain and spinal cord tumors. Since the launch of the project, OpenPBTA has collaboratively created reproducible workflows for integrated consensus SNV, CNV, and fusion calling, enabled RNA-Seq-based classification of medulloblastoma subtypes, and more than 25 additional DNA- and RNA-based analyses. The open-science platform and associated datasets and processed results provide a continuously updated, global view of the integrated cross-disease molecular landscape of pediatric brain tumors. Such biospecimen- and clinically-linked scalable data resources provide unprecedented collaborative opportunities for precision-based, personalized therapeutic discovery and drug development with the upcoming further integration of proteomic sample data (N >300) and drug response datasets, additionally diversifying the multimodal discovery potential of crowd-sourced approaches for accelerated impact for children with brain tumors.

Visualizing genomic characteristics across an RNA-Seq based reference landscape of normal and neoplastic brain

“…Constructing the Brain-UMAP / Clustering of gene expression data identifies diverse disease types. To characterize and better understand the molecular intricacies of brain tumors, we downloaded uniformly processed RNA-Seq abundance values from recount-brain, a curated repository for human brain RNA-Seq datasets, for three different uniformly processed datasets-702 adult glioma samples from TCGA 1 , 270 adult glioma samples from CGGA 5,6 , 1409 healthy normal brain samples from GTEx 4 across 12 GTEx-defined brain regions (Supplementary Table 1a). Retrieving data from recount 7 ensured that consistent bioinformatic pipelines were used for these three datasets thus resulting in no batch effects between the three datasets.…”

Multiparametric MRI Along with Machine Learning Informs on Molecular Underpinnings, Prognosis, and Treatment Response in Pediatric Low-Grade Glioma

“…To identify the enrichment of immune and stromal cell types and pathways, xCell 18 , which is a cell type enrichment analysis from gene expression data, was applied on our pLGG transcriptomic data. We leveraged transcripts per million (TPM) generated by a harmonized STAR-RSEM RNA-sequencing pipeline developed as part of the Gabriella Miller Kids First Data Resource Center sequencing and benchmarking effort 78 , which newly leverages Ensembl's GENCODE version 39 80 for gene-level annotations. We leveraged the Immunedeconv R package 81 specifying default parameters and xCell as the method of interest.…”

ALT in Pediatric High-Grade Gliomas Can Occur withoutATRXMutation and is Enriched in Patients with Pathogenic Germline MMR Variants

“…Here, we assess the frequency of ALT, as well as clinical and molecular phenotypes associated with ALT, in a large cohort of pediatric brain tumors from the OpenPBTA, with detailed investigation of HGATs [32][33][34] .…”

“…Here, we assess the frequency of ALT, as well as clinical and molecular phenotypes associated with ALT, in a large cohort of pediatric brain tumors from the OpenPBTA, with detailed investigation of HGATs 32-34 . We validated the use of the computational algorithm, TelomereHunter 19 , to predict ALT status.…”