inconsistency between percentage of duplicates in rnafusion

[fevac]

Description

• We are using two different metrics to report percentage of duplicates (calculated by different tools) in rnafusion. One is used to do the QC metrics checks and the other one is reported in the delivery file. These should be unified and we should only use one of them.

Clarification

When performing the QC check the duplication metric from one tool is used but when informing the customer of the quality in the delivery report the metric from another tool is reported.

To look up: which software produces which metric

[Vince-janv]

@fevac Which metric should be used for both the QC check and shown in the delivery report?

[ivadym]

So, we use pct_duplication to report metrics in the delivery report and PERCENT_DUPLICATION for QC validation. @rannick, could you provide some feedback on which one we should standardize on moving forward?

[rannick]

I think we need to unify which metric we use between tomte/mip-rna and rnafusion. At the moment rnafusion can get this metric from fastp and markduplicates. I think mip-rna reports markduplicates results and we should align rnafusion to that