kl-metapool beta.zip #141
kl-metapool beta.zip #141
Conversation
There was a problem hiding this comment.
Almost all of the changed files have the issue that it isn't sufficient to just set the SheetVersion to 90 without changing the name of the (currently) well_id_384 column to Sample_Well, which is what is required by v90. Most of the previous errors that are no longer occurring in this version were fixed because the setUp of various test classes is no longer failing on the load of good-sample-sheet.csv/good_sheet1.csv into internal variables, because these two DID get the Sample_Well fix :)
The four sheets associated with replicates (the original and the three demuxed versions) unfortunately CANNOT be v90, since that doesn't support replicate info. They will have to be put back to SheetVersion 100 and have contains_replicates put into their Bioinformatics project tables (True for all projects for the original file and False for all projects for the demuxed files).
| @@ -2,7 +2,7 @@ | |||
| IEMFileVersion,4,,,,,,,,,, | |||
| Investigator Name,Knight,,,,,,,,,, | |||
| SheetType,standard_metag,,,,,,,,,, | |||
| SheetVersion,100,,,,,,,,,, | |||
| SheetVersion,90,,,,,,,,,, | |||
There was a problem hiding this comment.
AFAICT, this sheet is never actually used in an actual test. I see that TestNuQCJob, in setUp, assigns this sample sheet's path to self.bad_sheet_bools_path ... but I can't find where self.bad_sheet_bools_path is used.
If it really isn't used anywhere, let's just delete it. If it IS, we will have to fix it (depending on what is supposed to be bad about it ...). We can't just change the version to 90 without also updating the well_id_384 column nane to Sample_Well.
| @@ -1,7 +1,7 @@ | |||
| [Header],,,,,,,,,,, | |||
| IEMFileVersion,4,,,,,,,,,, | |||
| SheetType,standard_metag,,,,,,,,,, | |||
| SheetVersion,100,,,,,,,,,, | |||
| SheetVersion,90,,,,,,,,,, | |||
There was a problem hiding this comment.
This sheet remains invalid because of the "v90 needs Sample_Well" issue ... but it doesn't really matter because it has an invalid assay value so it never gets that far, so the test it is associated with passes. I think we should probably make the Sample_Well change anyway, though, so that we have just one error in the file and don't have to worry about some dastardly order change ...
No description provided.