Read QC drops for individual plates into a merged dictionary, write it to YAML, and also convert to a DataFrame. If you really want to look into the details of what is being dropped, you will want to look at that merged YAML file.
Writing merged plate drops to results/qc_drops/plate_qc_drops.yml
No plate had any drops of the following types, so they are not plotted below: barcodes
Now plot the number of drops for each plate. A plate with no drops at all still gets a row, which is simply empty. You should be worried (maybe re-do or discard) any plates with a very large number of drops:
If a barcode is dropped especially often across plates, that could indicate something problematic with that barcode such that it should be removed altogether from the library analysis.
First, write a YAML with this information:
Writing merged barcode drops to results/qc_drops/barcode_qc_drops.yml
Now make a plot showing how often each barcode is dropped for each reason:
Analyze the QC performed on the groups/sera, which involves completely dropping titers for certain virus-sera pairs.
Read the QC for different groups/sera into a merged dictionary, write it to YAML, and also convert to a DataFrame. If you really want to look into the details of what is being dropped, you will want to look at that merged YAML file.
Writing merged groups/sera drops to results/qc_drops/groups_sera_qc_drops.yml
No virus titers were dropped at serum QC, so the two plots below are empty.
Plot the number of viruses dropped for each group/serum. If a group/serum has many missed viruses, then you will lack a lot of titers and so it may be worth reviewing the cause of the drops.
Plot the number of sera for which each virus is dropped during serum QC. If a virus is dropped for many sera, that may indicate some issue with that virus in assays: