Update percentile threshold behavior for pVACvector in 7.0

[susannasiebert]

pVACseq/bind/fuse/splice in 7.0 produce aggregated percentile metrics on three levels of granularity: binding, presentation, and immunogenicity. There are also three cutoffs --binding-percentile-threshold, --presentation-percentile-threshold, and --immunogenicity-percentile-threshold replacing the previous --percentile-threshold. A filtering/prioritizing on the combined percentile is no longer possible on those tools.

We need to decide how we want pVACvector to behave in this regard going forward. Should it continue using only the combined percentile? Or should it also have individual cutoffs? Moreover, the individual cutoffs now come with a default. Is that something we desire in pVACvector as well? Lastly, we need to discuss whether to incorporate normalized percentiles here as well.

[susannasiebert]

Discussion result from November 25th pVACtools meeting: For now, let's only using the binding percentile (instead of the combined percentile) and keep the default cutoff of 2.0.

No decision was made on using the normalized percentiles.