Release RCAS release v1.1.1 · BIMSBbioinfo/RCAS

New features

New function introduced: plotFeatureBoundaryCoverage that creates interative plotly graphs for coverage profiles at transcript feature boundaries
Changes in the HTML output of runReport function:
- The interactive plots now can be exported to not only PNG but also SVG for higher quality figures.
- The interactive tables for GO and GSEA analyses now have additional column for Benjamini-Hochberg multiple testing corrected p-values.
- The design of the coverage profiles have changed:
  - The coverage profiles are represented not only by a line but also a confidence envelope surrounding the line. The solid-line in the middle represents the mean coverage score (output of ScoreMatrix function from genomation package), and the thickness of the surrounding confidence envelope is as large as the standard error of the mean multiplied by 1.96 (95% confidence interval).
- The font sizes in all plots have been increased to 14.
calculateCoverageProfile function returns a Score Matrix object rather than a data.frame.
calculateCoverageProfileList object returns a data.frame rather than a list of data.frame objects.

In inst/report.Rmd script (called by the runReport function), the motif analysis chunk has been updated to account for the number of input intervals before down-sampling to 10000.
Coverage profile calculations for coding exon-intron junctions has changed:
- Now, firstly all internal exons are found such that all exons have at least two neighboring introns on both 5' and 3' directions. Then, the coverage profiles on the 5' and 3' directions are calculated. Previously, this filtering for the criteria that the exon must be in the middle of two introns was not applied and only a 50 bp region was plotted. Now, the profiled region is the area that spans 1000 bps both upstream and downstream of the internal exons.

Due to a bug fix in older version of genomation package, RCAS now depends on genomation releases later than 1.5.5.
plotrix library is added to dependencies. The std.error function from this library is use to calculate standard error of the mean of a given vector of real numbers.