Pages
Nederland, Amsterdam, Academisch Medisch Centrum
Contact details
Traineeship proposition
Abstract
Testimony
Admin
Stage-onderwerp 1 banaba bio-informatica 2015-2016: Evaluation of pseudo-alignment based RNA-Seq quantification
Stage-onderwerp 2 banaba bio-informatica 2015-2016: Breast cancer prognosis using H3K27ac ChIP-Seq data
Abstract 2018-2019: Interactive reporting and visualization of DNA methylation and gene expression time series data using Shiny
Multiomics experiments are increasingly commonplace in biomedical research and add layers of complexity to experimental design, data integration, and analysis. A problem is that it is difficult to visualize this omics data without losing the context. It is not convenient to visualize the results of a multiomics experiment in a static way because there is too much data, so this needs to be done in an interactive way. A typical example of a multiomics experiment consists of gene expression and DNA methylation data. Such experiments are often performed to better understand the (regulatory) relation between gene expression and the methylation status of the promoter and enhancer regions. For this research subject, breast cancer cell line data of two replicates was provided as a MultiAssayExperiment (mae) R object. This mae object contains gene expression data and DNA methylation data as different assays. To be able to link these types of data, the genes were annotated with their corresponding CpG sites. As mentioned before, the visualizations need to be interactive, so the R package Shiny was used. These types of data can be visualized both separately and in combination. Because a link was added between these types of data, the correlation between gene expression and DNA methylation can be calculated and plotted. To also visualize the data in its genomic context, a combined plot can be created using the R package Gviz.
Samenvatting eindwerk 1 2014-2015: Galaxy pipelines for omics data analysis
Increased reliance on computational approaches in the life sciences has revealed grave concerns about the accessibility and reproducibility of the obtained results. Galaxy (http://usegalaxy.org), an open web-based platform for genomic research, addresses these problems. Galaxy automatically tracks and manages data provenance and provides support for capturing the context and intent of computational methods.
The goal of this project is to contribute to ongoing efforts in the bioinformatics group at the Academisch Medisch Centrum (AMC) to make several analysis pipelines for high-throughput experiments with omics data available via the web-based platform Galaxy. Another aim of the project is to enable scientists without programming experience to easily specify parameters and run the Galaxy tools and workflows by themselves.
To achieve these goals the following methods were addressed during the internship: installing one’s own Galaxy environment, learning how to add tools to this Galaxy instance using Galaxy Tool XML (eXtensible Markup Language) files, getting the previously developed Galaxy tools from the bioinformatics group working again and implementing newly developed analysis tools within the Galaxy environment.
The outcome of this project was that the seven tools of the AMC are available again in Galaxy. The workflow for these Galaxy tools has also been restored and is working again. In addition, the workflow has been improved by creating a datatype that allows to upload Rdata objects, containing an ExpressionSet, at any point of the workflow. Above that, there was also created an own Galaxy tool that is available as well .
Although scientists don’t need programming experience to specify parameters and run tools in Galaxy, it would be useful for scientists, who want to modify an analysis to their needs, to learn a programming language such as R to adjust the tools.
Samenvatting eindwerk 2 2014-2015: Reporting and visualization of omics analysis results in interactive HTML reports
The aim of this research was to find a more interactive and easily to automate way to report and visualize the results of omics analyses. This reporting is necessary because the analyses must be easily shareable with other research groups who in general have less experience with bioinformatics and statistical programming languages such as R.
Currently, at the Bioinformatics Laboratory of the Academic Medical Center, Powerpoint is used to make a report of the analysis. This presentation contains all the relevant plots and links to reports from tools in R. The layout and content of this PowerPoint, however, strongly depends on the type of analysis and purpose of the underlying research. This way of presenting the resulting data is too static and time consuming. There is a need for a more interactive way of presenting the results. This way a researcher can easily change the parameters, variables and number of points in order to view a plot, or table exactly the way he wants. The produced report should also be easy reproducible. Automating the process of reporting would also make it more reproducible and with less effort.
There are several tools and R packages to make reactive and reproducible reports. But these tools aren’t made to create a full report. Most of the time they are specifically made for the creation of a specific part of the complete report.
Therefore, in the light of this research, an application has been developed which uses the web application framework ‘Shiny’ for R. With this application it should be possible to automate the process of making an omics related report. When the analysis is done, the results can be put in standard objects in R. These objects are used in the shiny-application. This way, a report is made with the values of these objects. By this way, the report is automatically generated. This report is also interactive and gives the information in an orderly manner.
Currently, at the Bioinformatics Laboratory of the Academic Medical Center, Powerpoint is used to make a report of the analysis. This presentation contains all the relevant plots and links to reports from tools in R. The layout and content of this PowerPoint, however, strongly depends on the type of analysis and purpose of the underlying research. This way of presenting the resulting data is too static and time consuming. There is a need for a more interactive way of presenting the results. This way a researcher can easily change the parameters, variables and number of points in order to view a plot, or table exactly the way he wants. The produced report should also be easy reproducible. Automating the process of reporting would also make it more reproducible and with less effort.
There are several tools and R packages to make reactive and reproducible reports. But these tools aren’t made to create a full report. Most of the time they are specifically made for the creation of a specific part of the complete report.
Therefore, in the light of this research, an application has been developed which uses the web application framework ‘Shiny’ for R. With this application it should be possible to automate the process of making an omics related report. When the analysis is done, the results can be put in standard objects in R. These objects are used in the shiny-application. This way, a report is made with the values of these objects. By this way, the report is automatically generated. This report is also interactive and gives the information in an orderly manner.
|
Tags: bioinformatics |
Address
|
Meibergdreef 9
1105 AZ Amsterdam (Nl)
Netherlands |
Contacts
|
Traineeship supervisor
Perry Moerland
|