search

instedd / cdx

Connected Diagnostics Platform
https://cdx.io
9 stars 7 forks source link

When uploading samples with a CSV, the replicate number should be taken as an input #1972

Closed diegoliberman closed 1 year ago

diegoliberman commented 1 year ago

If we don't allow to enter the replicate number, and if they input a CSV with samples from the same batch and the same concentration, all of them remain with the same replicate number --> 1

I think that a column in the CSV should be "replicates". If the user enters "4", then the system will create 4 replicates for that row.
An alternative would be that the replicate column has the actual number so the user has full control. I'm not sure what would be more comfortable for them. Maybe we could ask AlexC?

ysbaddaden commented 1 year ago

I put the question label to first verify & figure out how the CSV import is working.

ysbaddaden commented 1 year ago

Behavior confirmed: one line in the CSV is one generated sample;. This results in identical samples being marked as replicate 1, when the other methods will create identical samples marked as replicate 1, 2, 3, ... The behavior also differs from other methods, where we select a number of replicates.

Here are some potential solutions:

  1. add a replicate column to the CSV with the actual number (:-1: annoying);
  2. add a replicates column to create N replicates from one definition (:+1 similar to other methods);
  3. identify replicated columns in the CSV and group them (:shrug: why not);
sardar-usman commented 1 year ago

@leandroradusky It is working as expected.