GitLab

- `cp_test` was randomly selecting 1500 genes if the input was larger than this number
- this lead to irreproducibility
- fix was to remove this line, and add the fixed function to the script, replacing the sourced code

Fixes #98

- for very small values, < 2.225074e-308,  `R` will round to zero because
they are bellow the precision point
- zeros lead to Inf values when calculating the Rank values which error
- solution was to replace zeros with a very small value, the lowest
precision point

- new table for the size of input gene sets
- new table with the number of non-mapped genes (bitr) for each contrast

closes #97

- increased the size outer ring. the color represents the category and
was not very visible.

- ranks were taken from log2fc instead of rank_score

Fixes #96

- removed extra dev.off

Part of #96

- loops over contrasts
- added test arguments for set size
- saves fgsea graphical table and sources it to markdown
- results table contains contrasts

Closes #96

- added new arguments for set size use

closes #95

- detects species from gene names for gene name conversion.
Species is still needed as an input to retrieve the sets from GSEA

Fixes #94

- The logical test to check if there was at least one gene set input was always defaulting to TRUE.
- List of genes is created even if the --gmt_file is the only option set

- flexible enough that any of --gmt_file, --extra_sets, or --category can be used
- fixed minor typo in help
- styled script

See #93

- added several test to confirm list selection
- it is now possible to use only custom gene list. Before they were always added to an Hallmark gene list.
- changed variable name to fit coding style
- added new plot for an overview of enriched sets

Closes #92

Closes #92

enabled handling of 'description' and 'gene_description'; adjusted gather for inclusive rather than exclusive handling of column selection; removed port information

- extracts stranded counts using a argument

Fixes #88

- moved to section 1.7

- Added text explaining the goal

- dot plot now shows alpha using difference of non-exonic counts

Features for #51

- added new arguments to choose GTF featrue for genes and exons
- implemented attributes and features
- added test to explain the issue of overlapping genes
- fixed missing  knitr::kable call
- added overlapping genes color to dot plot

- added min_count argument to avoid overplotting
- added correlation plots

Development story can be seen here:
https://git.mpi-cbg.de/scicomp/bioinfo_team/test_rna-seqc/commits/master/genic_counts.R

And it was originated in #51

- now alpha takes the value of `qcutoff` if set
- if `qcutoff` is not set, then it defaults to 0.1
- fixed typo in argument description

Fixes #23