Execution workflow: APAtrap - report Format 04 BED file in workflow (per-PAS fractional relative usage)

[faricazjj]

Parent issue - #382

Disclaimer
The implementation in this PR has been discussed in the POC to calculate relative usage in tools #278

Background
APAtrap's original absolute TPM quantification output for control_r1 looks like the following:

For each transcript id, we have a list of all identified PAS and their expression levels from the most proximal to most distal if the strand is +, and the opposite for a - strand (source: APAtrap wiki).

Implementing relative usage calculation and output
The raw output above is then transformed into the following quantification output file:

To get relative usage output, we would get each PAS in the transcript divided by the total sum of expression levels of all the PAS in the same transcript. For example, the screenshot above of the original output would be converted to the following:

Taking the example of ENSMUST00000023572.14, from the quantification output file, we get the following:

ENSMUST00000023572.14|16|78336188-78340752|+    0.0
ENSMUST00000023572.14|16|78336188-78340752|+    50.15
ENSMUST00000023572.14|16|78336188-78340752|+    35.57   

In the relative quantification output file, the score column of the corresponding rows become:

ENSMUST00000023572.14|16|78336188-78340752|+    0.0
ENSMUST00000023572.14|16|78336188-78340752|+    0.5850443303779748 
ENSMUST00000023572.14|16|78336188-78340752|+    0.4149556696220252