New MiSeq and NextSeq error models

[StefanLelieveld]

Hi Hadrien,

I have added the new MiSeq and NextSeq models to the profiles-folder and made them available through the cli. Both the MiSeq and NextSeq error models produce paired-end reads of 301bp. The miSeq models are based on the reads of the 8 runs of SRA project ERP022313.

• miSeq_0.npz --> filter: none
• miSeq_20.npz --> filter: the forward and reverse read have a mean quality score >= 20
• miSeq_24.npz --> filter: the forward and reverse read have a mean quality score >= 24
• miSeq_28.npz --> filter: the forward and reverse read have a mean quality score >= 28
• miSeq_32.npz --> filter: the forward and reverse read have a mean quality score >= 32
• miSeq_36.npz --> filter: the forward and reverse read have a mean quality score >= 36
• nextSeq.npz --> filter: none

Below are the placeholder-names that I have used for the error models in the cli. I am not very happy with the current names: it's not clear what the number means. At the same time I don't want to use very long and extensive names. The naming of the error models is open for discussion.

• miSeq_0.npz --> "miseq"
• miSeq_20.npz --> "miseq-20"
• miSeq_24.npz --> "miseq-24"
• miSeq_28.npz --> "miseq-28"
• miSeq_32.npz --> "miseq-32"
• miSeq_36.npz --> "miseq-36"
• nextSeq.npz --> "nextseq"

Let me know what you think of it.

Best,
Stefan

[HadrienG]

I've been thinking about the model names and I don't think I have a much better naming scheme to propose.

Eventually could a q to make it a bit more obvious it's about the quality scores: miseq-q20 or miseq-q20+, but leaving them as is would be fine with me too.

[HadrienG]

Are we happy with the names and merging this?