Region Of Interest example
Example showing some of the features of the ROI (region of interest) concept.
The idea
The basic idea of ROIs is to decouple a genomic region from it genomic context.
This allows to calculate/define "features" or metrics for these regions. Thus, they can be sorted. Doing so, we can show all regions with a specific property in close proximity to each other. This is important when trying to understand for example coverage data.
Previously one would have to scroll through the genome using a genome browser (e.g. IGV) to do so. (Or write a program ...)
How it works
A ROI file that was generated using the SEQAN tools (http://www.seqan.de/projects/ngs-roi/) is being read by ROIReader. We then remove regions that shorter than 100 and have less 100 counts. Sorting by nac (normalized average coverage) allows us to view "peaky" regions in close proximity. The IGVview node allows to link the regions to the IGV genome browser.
As can be seen from the CollectionLinePlot view we have regions with large peaks and uneven coverage at the top of the list...
Case Studies: Support of Target Discovery Research by KNIME Textprocessing and Networkminig
