A Comprehensive Infrastructure for Big Data in Cancer Research: Accelerating Cancer Research and Precision Medicine

Abstract

Advancements in next-generation sequencing and other -omics technologies are accelerating the detailed molecular characterization of individual patient tumors, and driving the evolution of precision medicine. Cancer is no longer considered a single disease, but rather, a diverse array of diseases wherein each patient has a unique collection of germline variants and somatic mutations. Molecular profiling of patient-derived samples has led to a data explosion that could help us understand the contributions of environment and germline to risk, therapeutic response, and outcome. To maximize the value of these data, an interdisciplinary approach is paramount. The National Cancer Institute (NCI) has initiated multiple projects to characterize tumor samples using multi-omic approaches. These projects harness the expertise of clinicians, biologists, computer scientists, and software engineers to investigate cancer biology and therapeutic response in multidisciplinary teams. Petabytes of cancer genomic, transcriptomic, epigenomic, proteomic, and imaging data have been generated by these projects. To address the data analysis challenges associated with these large datasets, the NCI has sponsored the development of the Genomic Data Commons (GDC) and three Cloud Resources. The GDC ensures data and metadata quality, ingests and harmonizes genomic data, and securely redistributes the data. During its pilot phase, the Cloud Resources tested multiple cloud-based approaches for enhancing data access, collaboration, computational scalability, resource democratization, and reproducibility. These NCI-led efforts are continuously being refined to better support open data practices and precision oncology, and to serve as building blocks of the NCI Cancer Research Data Commons.

Predicting the outcome of patients with chronic lymphocytic leukemia: Progress and uncertainty

“…Will next-generation sequencing, "omics," liquid biopsies, big data, and machine learning identify more accurate and precise biomarkers? 16,17,58,59 The answer is hopefully yes. However, inaccuracy and imprecision cannot be eliminated entirely because of measurement error, chance, and uncertainty.…”

“…However, inaccuracy and imprecision cannot be eliminated entirely because of measurement error, chance, and uncertainty. 1,[59][60][61][62] Analyses based on new technologies applied to large data sets (big data) have downsides, and may in fact demand a greater tolerance for uncertainty. 60,61 On the other hand, these analyses are hypothesis generating, not hypothesis testing.…”

Precision Medicine in Oncology: In Vitro Drug Sensitivity and Resistance Test (DSRT) for Selection of Personalized Anticancer Therapy

“…For example, presence of particular Single Nucleotide Polymorphisms (SNP) in genome can predict response of a patient to a therapy . Numerous projects have been focused on collecting databases of “omic” data, as well as software tools, and making them available for public use . Examples of biggest projects are Cancer Genome Atlas (TCGA), collecting genetic mutations causing cancer; Human Proteome Project (HPP) documenting all of the proteins of healthy human body; and International Human Epigenome Consortium (IHEC) generating human epigenomes from different types of healthy and disease‐related human cells.…”

Analyses of cancer data in the Genomic Data Commons Data Portal with new functionalities in the TCGAbiolinks R/Bioconductor package

“…In this context, numerous large-scale studies have been conducted using state-of-the-art genome analysis technologies. One of the most important examples is The Cancer Genome Atlas (TCGA), which started in 2006 as a pilot project aiming to collect and conduct analyses on an unprecedented amount of clinical and molecular data including over 33 tumor types spanning over 11,000 patients, subsequently generating more than 2.5 petabytes of publicly available data over the past decade [8–10]. Publicly funded by The National Institute of Health (NIH), TCGA has made numerous discoveries regarding genomic and epigenomic alterations that are candidate drivers for cancer development, and this was achieved through creating an “atlas” and applying large-scale genome-wide sequencing and multidimensional analyses.…”

“…Publicly funded by The National Institute of Health (NIH), TCGA has made numerous discoveries regarding genomic and epigenomic alterations that are candidate drivers for cancer development, and this was achieved through creating an “atlas” and applying large-scale genome-wide sequencing and multidimensional analyses. These latter efforts have significantly contributed to high-quality oncology studies, either led by the TCGA research network or other independent researchers [10], which recently culminated in 27 original publications from the PanCancer TCGA initiative [11]. In 2016, TCGA was moved under the umbrella of the broader repository Genomic Data Commons (GDC) Data Portal [12] together with other studies.…”