Add info about the sequencing tech TAG and reflect that on the reports

[abhi18av]

As part of 4-APR meeting.

• Focus on homogenous (sequencing) datasets
• (IN FUTURE) Accommodate hybrid datasets and reflect on the final results (nanopore/illumina)

@vrennie @TimHHH , where exactly do we need to add this sequencing platform information i.e. which summary files?

[TimHHH]

@vrennie @TimHHH , where exactly do we need to add this sequencing platform information i.e. which summary files?

I would think a column in the summary stats file.

[vrennie]

Yes, I agree with Tim, just a column that looks like this:

Sequencing Technology Illlumina ONT ONT Illumina Illumina Illumina ...

[abhi18av]

Okay, I understand this would be added to the summary stats file 👍

However, there's one more detail worth mentioning here, currently we hard-code the sequencing technology in the bam_rg_string https://github.com/TORCH-Consortium/MAGMA/blob/786d13dfe1988784f870499bb878f47cc945a493/workflows/validate_fastqs_wf.nf#L30

Should we not add this column to the input-samplesheet as well?

[vrennie]

Yes, good catch @abhi18av, lets add this as a column to the samplesheet.

[TimHHH]

Yes, ideally the user provides the sequencing technology in the sample sheet and this is then used in the bam_rg_string along the lines of PL:${technology}. The documentation has to be clear that only one technology is allowed per sample sheet.

[abhi18av]

Guys, what about reflecting that on the actual sample name as well? Something like Shea2017_2021_396.SRR16089406.LNA.A1.ILMN.1.1.1

The NCBI currently lists the following platforms used for the sequences

• ILLUMINA
• ION_TORRENT
• ABI_SOLID
• PACBIO_SMRT
• CAPILLARY
• OXFORD_NANOPORE
• LS454
• BGISEQ

To avoid long names, we can perhaps standardize the acronyms like ILMN / ONT / PCB / ION etc - what do you think?

[vrennie]

I think unless the full name messes up the .csv its better to keep the full name