bailey-lab / mipscripts Goto Github PK

During the merge_sampleset subcommand, we merge together multiple sample sheets by column names. However, when column names do not align, this can cause issues.

We recommend the use of snake case as a standard in naming column names. However, there are situations where this may not occur and the capture plate name column may appear as capture_plate_name in one sample sheet and Capture Plate Name in another sample sheet.

To address this, we should standardize file column names when files are fed into mipscripts functions.

If column names are duplicated within a sample sheet, this can cause problems when we try to merge sheets or even when we compute stats on the sequenced data.

It would be ideal to throw an error when this occurs to inform the user of the issue.

During the seqrun_stats subcommand, we check if FASTQ names for a particular sample name, sample set, and replicate are duplicated. In some cases, we flag FASTQs as duplicates even though they are not. For example,

python3 -m mipscripts seqrun_stats --samplesheet /nfs/jbailey5/baileyweb/bailey_share/raw_data/220518_nextseq/220518_samples.tsv

returns an error for the FASTQs: 36DA-EPHI-1_S22_R1_001.fastq.gz and 1836DA-EPHI-1_S92_R1_001.fastq.gz. Here seqrun_stats is not able to discriminate between the sample names (36DA and 1836DA).