run file read_type column stopping submission

[handibles]

Thanks for the efforts involved in this package.

With the following for fastq, and in a fresh conda env for this package only:

ena-upload-cli \
  --action add \
  --center 'Funtime Sequencing Inc.' \
  --checklist ERC000011 \
  --no_data_upload \
  --data *gz \
  --secret ~/.webin.yaml \
  --xlsx /path/to/excel.xlsx

Hitting a persistent issue using the template XLSX file, where a read_type column appears mandatory (not optional). Without adding the read_type column, upload issues the following:

Check if all required columns are present in the study table.
Check if all required columns are present in the experiment table.
Check if all required columns are present in the run table.
Check if all required columns are present in the sample table.
No valid checksums found, generate now... done.
No files will be uploaded, remove `--no_data_upload' argument to perform upload.
Retrieving taxon IDs and scientific names if needed
Taxon IDs and scientific names are retrieved
Constructing XML for 'study' schema
wrote /tmp/study_edf0d8d4-1528-47a7-9b6e-b3d3dd364b3c.xml
Constructing XML for 'experiment' schema
wrote /tmp/experiment_c6a05995-d247-4c06-8121-06dd6a5a65f1.xml
Constructing XML for 'run' schema
wrote /tmp/run_5097420a-f064-4d57-9ea7-44e5eb16fbd2.xml
Constructing XML for 'sample' schema
wrote /tmp/sample_75de0545-1f36-45c7-9a06-d2ed4f1eb7b6.xml
Constructing XML for submission schema
wrote /tmp/submission_ca88559c-2475-4305-ac8f-f7b3b3152aee.xml

Submitting XMLs to ENA server: https://www.ebi.ac.uk/ena/submit/drop-box/submit/
Printing receipt to ./receipt.xml

Oops:
In run, alias: "prenatal-PBX-affects-offspring__16S_run". Read type information missing in run.

I've added a column to the run sheet in the excel.xlsx template as below, using the controlled vocab linked in the readme.md, i.e. a column reading "paired" :

alias | experiment_alias | file_name | file_format | read_type
(Mandatory)   Unique identificator for each run. | (Mandatory)   From_experiment_metadata | (Mandatory) The name or relative   pathname of a run data file. | (Mandatory) The run data file   model. |  
prenatal-PBX-affects-offspring__16S_run | prenatal-PBX-affects-offspring__16S_amplicon | prenatal-PBX-affects-offspring__11_1_S115_R1.fastq.gz | fastq | paired
prenatal-PBX-affects-offspring__16S_run | prenatal-PBX-affects-offspring__16S_amplicon | prenatal-PBX-affects-offspring__11_1_S115_R2.fastq.gz | fastq | paired
prenatal-PBX-affects-offspring__16S_run | prenatal-PBX-affects-offspring__16S_amplicon | prenatal-PBX-affects-offspring__11_2_S129_R1.fastq.gz | fastq | paired
prenatal-PBX-affects-offspring__16S_run | prenatal-PBX-affects-offspring__16S_amplicon | prenatal-PBX-affects-offspring__11_2_S129_R2.fastq.gz | fastq | paired

Which then generates the following oopsception:

[ ... as above ]

Oops:
In run, alias: "prenatal-PBX-affects-offspring__16S_run". Read type configuration in run is invalid.

Thanks in advance for the help. Happy to try mock up a minrepex if helpful.
[edit: removed duplicated --checklist in the CLI, not related]

[bedroesb]

@handibles Thanks a lot for the detailed report. Is this single cell data? I am not familiar myself in submitting more than 2 files per experiment/run alias (and thus not needing the read_type), but based on the PR implementing this feature: #102, you seem to be doing things correctly. @cgirardot sorry tagging you into this, but do you happen to see what is wrong?

[handibles]

Thanks @bedroesb. Ha no, this is more or less the opposite - 188 paired FASTQ from environmental DNA / microbiome. I've attached a paired-back version of the xlsx used for reference.

microbiome.xlsx