The ExAC browser: displaying reference data information from over 60 000 exomes.

Karczewski KJ, Weisburd B, Thomas B, Solomonson M, Ruderfer DM, Kavanagh D, Hamamsy T, Lek M, Samocha KE, Cummings BB, Birnbaum D, The Exome Aggregation Consortium, Daly MJ, MacArthur DG. · 2017

Read The ExAC browser: displaying reference data information from over 60 000 exomes. by Karczewski KJ, Weisburd B, Thomas B, Solomonson M, Ruderfer DM, Kavanagh D, Hamamsy T, Lek M, Samocha KE, Cummings BB, Birnbaum D, The Exome Aggregation Consortium, Daly MJ, MacArthur DG. (2017) in the Tower of Babel, a free online library of public-domain works preserved in clean, searchable text. No sign-up. The full text is served as a plain-text file at a predictable URL, where text is present: https://antarctic-labs.com/tower-of-babel/europmc-pmc5210650.txt.

Introduction

Konrad J. Karczewski and colleagues, with the Exome Aggregation Consortium, presented the ExAC browser in Nucleic Acids Research in 2017. Knowing how common a gene variant is in the general population helps clinicians judge whether it causes disease or is harmless. The browser displays reference variation data drawn from the exomes, the protein-coding parts of the genome, of 60,706 people.

Views are organized by gene and by transcript, which suits clinical work, and a variant view adds population frequency, functional notes, and the short reads that support each call. The software is light, flexible, and open source, and any group can reuse the framework for other large population data sets. Clinical laboratories around the world already use it to weigh variants found in patients against this broad reference.

Catalog record

CREATORKarczewski KJ, Weisburd B, Thomas B, Solomonson M, Ruderfer DM, Kavanagh D, Hamamsy T, Lek M, Samocha KE, Cummings BB, Birnbaum D, The Exome Aggregation Consortium, Daly MJ, MacArthur DG.
YEAR2017
COLLECTIONPAPERS
FORMATTXT
FILE SIZE19 KB
LICENSECC BY 4.0
RIGHTSCC BY
SOURCEEurope PMC

Source: https://europepmc.org/articles/PMC5210650

Browse the full library · About the Tower of Babel

Full text

The opening of the work, from the library's hosted text:

The ExAC browser: displaying reference data information from over 60 000 exomes By Konrad J. Karczewski, Ben Weisburd, Brett Thomas, Matthew Solomonson, Douglas M. Ruderfer, David Kavanagh, Tymor Hamamsy, Monkol Lek, Kaitlin E. Samocha, Beryl B. Cummings, Daniel Birnbaum, Mark J. Daly Published in: Nucleic Acids Research DOI: 10.1093/nar/gkw971 ABSTRACT Worldwide, hundreds of thousands of humans have had their genomes or exomes sequenced, and access to the resulting data sets can provide valuable information for variant interpretation and understanding gene function. Here, we present a lightweight, flexible browser framework to display large population datasets of genetic variation. We demonstrate its use for exome sequence data from 60 706 individuals in the Exome Aggregation Consortium (ExAC). The ExAC browser provides gene- and transcript-centric displays of variation, a critical view for clinical applications. Additionally, we provide a variant display, which includes population frequency and functional annotation data as well as short read support for the called variant. This browser is open-source, freely available at http://exac.broadinstitute.org, and has already been used extensively by clinical laboratories worldwide. INTRODUCTION Recently, large reference datasets, such as those from the 1000 Genomes Project Consortium (1), Exome Sequencing Project (ESP) (2) and Exome Aggregation Consortium (ExAC) (3), have become publicly available for the benefit of the biomedical community. These datasets are beneficial for many applications, including clinical as well as basic research. In particular, as a variant's frequency is among the best predictor of its deleteriousness, clinical geneticists use reference datasets to discern pathogenic mutations from benign polymorphisms. Additionally, genetics researchers rely on variant and allele frequency data to infer gene and variant function (e.g. whether a gene is essential) as well as for population genetics analyses. These large-scale projects release raw data in the form of variant call format (VCF) files, but these files require bioinformatics expertise to parse and synthesize. Genome browsers, such as the UCSC genome browser (4), have become a popular method for non-technical audiences to visualize large genome-scale datasets. Additionally, browsers of variation data, including the Exome Variant Server (EVS) from ESP and the 1000 Genomes Browser, have been developed to present population data, but these are limited in the data they display. For instance, deviations in coverage, which affect one's confidence of the absence of variation, are not natively shown: EVS contains a link to a coverage track on UCSC, but coverage is not visualized on the page itself. There are a number of practical considerations for the optimal display of reference data. Specifically, as one primary use case for genome browsers involves gene-level analyses, the display of gene summary information is a central view for a genome browser, including integration of summary statistics as well as data for individual single nucleotide variants (SNV), insertions and deletions (indel) and copy number variants (CNV). Of course, detailed information on each variant, including annotations and quality metrics, is of paramount importance. However, an equally important display is that of the absence of variation: whether a missing variant implies a lack of observed variation, low or no coverage in the genomic region, or variation filtered due to poor quality. The Exome Aggregation Consortium (ExAC) has collected, harmonized, and released exome sequence data from 60 706 individuals (3). Already, these data have proven useful in filtering variants for identifying causal variants for rare disease (5,6,7). Here, we present a visual browser of the ExAC dataset. The browser is intended for use by clinical geneticists researching variants of interest for patients as well as biologists exploring variation in specific genes. EXAC BROWSER We designed the ExAC browser as an intuitive interface to enable clinical geneticists and biologists to explore variants and genes of interest. We built a scalable browser framework to display qualitative and quantitative information for genes and variants in the ExAC dataset (see Methods), including both quality control information as well as summary statistics. The front page of the browser includes a search bar, which is seeded with autocomplete suggestions based on gene symbols and aliases, as well as sample queries. From here, there are two central units of the ExAC browser: the gene (or transcript) page and the variant page. Gene/transcript page The ExAC browser gene page is an overview page for gene-level information, including summary statistics, coverage, and variants. The page begins with gene metadata and external references, along with constraint information, which summarizes the gene's intolerance to variation for multiple functional classes (3,8) (Figure 1A). Next, we present single base-resolution coverage information for each exon for a number of metrics including mean, median, and proportion of individuals covered at a number of depth cutoffs (Figure 1B). Immediately below, an exon summary plot displays the position and frequency of each SNV and indel, as well as CNV count information broken down by population (Figure 1C). All individual CNV calls are provided in the form of UCSC tracks, which are linked at the top of the page and above the CNV display. Finally, the browser provides a comprehensive table for variant information, which includes the worst functional annotation across transcripts for each variant, as well as frequency information. The table is sortable and can be exported to a CSV format (Figure 1D). [Figure: Gene page. (A) Gene information is summarized, including links to various external resources, as well as constraint information as described in (3). For all exons in the canonical transcript, we display (B) base-level coverage for a number of metrics (mean coverage by default), as well as (C) position and frequency information for all variants, including CNVs. (D) A table of all variants is provided with additional annotation information and links to variant pages.] By default, the gene summary page presents a table of all variants in the gene, annotated with the worst consequence across all transcripts, as well as coverage information for the canonical transcript: we also present a transcript page that includes annotation and coverage information specific to that transcript. Variant page The variant page includes a diverse set of annotations for the given variant. First, a site overview (Figure 2A) and site- and genotype-level quality metrics are provided (Figure 2B). The user is notified whether any individual has another variant in the same codon (suggesting a multi-nucleotide variant, or MNV), whether the site is multi-allelic, or if a low number of individuals is covered at this locus. Functional annotations of the variant against each transcript including PolyPhen2 (9), SIFT (10), and LOFTEE annotations, as well as a sortable table of population frequencies are provided (Figure 2C). Finally, for users that wish to evaluate the validity of specific variants, raw short-read data from a subset of individuals is available for each variant. We provide an IGVweb visualization of the read pileup of a 125 bp window around the variant (Figure 2D) for a random sample of individuals with each variant, as well as a sampling of homozygous individuals, if available. For the first time among genome browsers, we provide users with a mechanism to efficiently visualize the raw read support for a variant and make assessments of its quality that may not have been detected by variant calling algorithms. [Figure: Variant page. (A) Variant metadata is displayed, including links to dbSNP, UCSC and Clinvar. (B) Users can browse quality metrics

Read the complete text