Web-TCGA: an online platform for integrated analysis of molecular cancer data sets - PubMed (original) (raw)

Web-TCGA: an online platform for integrated analysis of molecular cancer data sets

Mario Deng et al. BMC Bioinformatics. 2016.

Abstract

Background: The Cancer Genome Atlas (TCGA) is a pool of molecular data sets publicly accessible and freely available to cancer researchers anywhere around the world. However, wide spread use is limited since an advanced knowledge of statistics and statistical software is required.

Results: In order to improve accessibility we created Web-TCGA, a web based, freely accessible online tool, which can also be run in a private instance, for integrated analysis of molecular cancer data sets provided by TCGA. In contrast to already available tools, Web-TCGA utilizes different methods for analysis and visualization of TCGA data, allowing users to generate global molecular profiles across different cancer entities simultaneously. In addition to global molecular profiles, Web-TCGA offers highly detailed gene and tumor entity centric analysis by providing interactive tables and views.

Conclusions: As a supplement to other already available tools, such as cBioPortal (Sci Signal 6:pl1, 2013, Cancer Discov 2:401-4, 2012), Web-TCGA is offering an analysis service, which does not require any installation or configuration, for molecular data sets available at the TCGA. Individual processing requests (queries) are generated by the user for mutation, methylation, expression and copy number variation (CNV) analyses. The user can focus analyses on results from single genes and cancer entities or perform a global analysis (multiple cancer entities and genes simultaneously).

PubMed Disclaimer

Figures

Fig. 1

Fig. 1

Illustration of mutational data. a Provides an overview of the global mutational profile of TP53, TSHZ3 and VHL highlighting the cancer entity specific occurrence of mutations within BRCA and KIRC, while TSHZ3, as control, remains undetected. In b and c, the mutational profile is broken down into the type of predicted impact on the protein sequences. The slice size denotes the proportion of a mutation type in relation to all mutations. The given percentage denotes the proportion of a mutation type in relation to all tumor samples of a given entity

Fig. 2

Fig. 2

Illustration of methylation data. Different to all other analyses types for methylation data the global view is just provided for one cancer entity at a time, due to the fact that the gene is partitioned into different regions. a Provides a global profile of the methylation status in SFRP1 and SFRP4 in COAD, where SFRP1s is highly differential methylated in its body region and SFRP4, as control, remains mostly unmethylated. Further, b and c give an estimate of the differential methylation distribution, which also allows the detection of shifts. The x-axis shows difference of the ß-values of the methylation status between tumor and normal samples. The y-axis reflects the number of samples

Fig. 3

Fig. 3

Illustration of expression data. To examine the global expression profile, the genes expression status in multiple cancer entities is rendered in the same way, as in Fig. 1a. Further, a waterfall plot, as shown in (a), is rendered for each cancer entity including the genes KRAS, EGFR and TTF1, outlining the patients exceeding a user given threshold for the genes expression on the x-axis, with the expression level on the y-axis. This plot is available in an entity (shown here) or a gene centric view. Further, Web-TCGA provides entity wise box plots (b), depicting an overview for the expression status of a cancer entity and enabling the identification of the expression patterns per cancer entity

Fig. 4

Fig. 4

Illustration of CNV data. Just as for all other analyses types a global CNV profile is provided by Web-TCGA, as shown in a where the copy number status for the genes FGFR1 and PIK3CA within the entities LUSC and LUAD is shown, giving a first overview of the CNV profile, which is conducted of all samples harboring a gain or a loss. Furthermore, this profile can be created for all combinations of gains and losses alone, to distinguish the proportions of low and high level gains and losses. To further resolve these global profiles, Web-TCGA provides bar plots, as shown in b and c, for each entity chosen. These bar plots can be used to estimate the proportions of gains and losses within each entity in a much finer grain as in (a)

References

    1. Analysis-ready standardized TCGA data from Broad GDAC Firehose stddata__2014_09_02 run. [http://gdac.broadinstitute.org/runs/stddata__2014_09_02/data]. Accessed 01 Feb 2016.
    1. Zhu Y, Qiu P, Ji Y. TCGA-assembler: open-source software for retrieving and processing TCGA data. Nat Methods. 2014;11:599–600. doi: 10.1038/nmeth.2956. - DOI - PMC - PubMed
    1. Broad Institutes Confluence Wiki. [https://confluence.broadinstitute.org/display/GDAC/Download]. Accessed 01 Feb 2016.
    1. Mermel CH, Schumacher SE, Hill B, Meyerson ML, Beroukhim R, Getz G. GISTIC2. 0 facilitates sensitive and confident localization of the targets of focal somatic copy-number alteration in human cancers. Genome Biol. 2011;12:R41. doi: 10.1186/gb-2011-12-4-r41. - DOI - PMC - PubMed
    1. Li B, Dewey CN. RSEM: accurate transcript quantification from RNA-Seq data with or without a reference genome. BMC Bioinformatics. 2011;12:323. doi: 10.1186/1471-2105-12-323. - DOI - PMC - PubMed

Publication types

MeSH terms

LinkOut - more resources