Instructions

1. There are three scripts to be executed sequentially: 1) sfari-consolidation.sh, 2) continue-sample-swap.sh, and 3) bam-reheader.sh. The output of each script will contain files with such pattern: *-data-files-to-move_part-*
2. Please make sure you have permissions to 1) make directory in the working directory 2) rename files
3. Please make sure this command works: echo "hi" > greeting.txt && sed -i 's/hi/bye/g' greeting.txt. I checked the manual of sed and it should work without appending after the -i option but just to be safe.
4. Make sure none of these files *-data-files-to-move_part-* exist when you execute the script- otherwise we're just going to append to it.

To execute

Please execute the output files in order by their numbers.

Move left-behind (unique) into primary & get md5 sums for select samples

This script will generate 6 files: going from 1 to 6.

sfari-consolidation.sh 2> stderr.log

⚠️ Inspect & execute sequentially the files: 1 to 6.

Sample-swap: put select files in their rightful place

This script will generate 3 files: going from 7 to 9.

continue-sample-swap.sh

⚠️ Inspect & execute sequentially the files: 7 to 9.

Reheader select BAMs

This script will generate 2 files: going from 10 to 11.

bam-reheader.sh

⚠️ Inspect & execute sequentially the files: 10 to 11.

To make sure overwriting doesn't require the --force argument, try this test first: echo hi > hi.txt && echo bye > bye.txt && mv bye.txt hi.txt. The resulting hi.txt file should contain bye. 🎉

Things to check for

1. It would be weird to find "fast5.md" in -data-files-to-move_part-5.txt" but alas, go ahead and regenerate the fast5. 😩
2. Please debug if you find Unaccounted for scenario in creating md5 or Multiple md5 identified in stderr.log. You should not find multiple md5 files for any deepest directory.
3. If we're moving files, do the source path exist? Does the parent directory of the destination exist?

Discussion

1. For the sample swap, do you think it's necessary to samtools reheader where it applies? Precisely, the nanopore unmapped bams for those files. We can either remove the tags with misleading sample or perform regex in-placement.

Post-discussion

We are doing samtools reheader.

1. According to the samtools docs for reheader, the in-place replacement only works for CRAM files. The solution in bam-reheader script will perform the following steps:
   1. Perform the re-header & output to a different BAM file: *-data-files-to-move_part-10.txt"
   2. Overwrite the original file with the updated one: *-data-files-to-move_part-11.txt"