add subsample_reads #138
add subsample_reads #138
Conversation
AmandaBirmingham
left a comment
AmandaBirmingham
left a comment
There was a problem hiding this comment.
One request for change to an error message, a couple of questions
src/qp_klp/Protocol.py
Outdated
| f'| gzip > {f}') | ||
| _, se, rv = system_call(cmd) | ||
| if rv != 0 or se: | ||
| raise ValueError(f'Error during mv: {cmd}. {se}') |
There was a problem hiding this comment.
This error message seems misleading since it (still) says "Error during mv"; I think it should be changed to say it is reporting on errors that occur during seqtk.
src/qp_klp/Protocol.py
Outdated
| for f in files: | ||
| dn = dirname(f) | ||
| bn = basename(f) | ||
| nbn = join(dn, bn.replace('fastq.gz', 'subsampled.gz')) |
There was a problem hiding this comment.
Is it going to cause any problems later that the subsampled reads file doesn't end with 'fastq.gz'? I'm not familiar with the workflow but I know I've seen regexes around that expect this suffix ...
There was a problem hiding this comment.
I wanted to keep in the working directory a backup of the original file, just in case we need to debug it. This basically moves the original problematic file to a new name with subsample,, then in the next command the subsample will create a new (smaller) file with the name of the original. In fact, I'm relying on those regex to ignore the subsampled.gz. I'll add a comment about this.
There was a problem hiding this comment.
Ah, that makes sense :) However, I wonder if we could name them something other than "subsampled.gz"? That name makes me think that the file with that name IS the one that was subsampled, rather than being the original one. Could we name it "not_subsampled.gz" or something?
There was a problem hiding this comment.
Yes, while writing the comment I realized the same thing so I called it "full".
| self.convert_raw_to_fastq() | ||
| self.integrate_results() | ||
| self.generate_sequence_counts() | ||
| self.subsample_reads() |
There was a problem hiding this comment.
Does this mean that we will now ALWAYS subsample?
There was a problem hiding this comment.
Yes, but really we will always check if subsample is needed and only run it when necessary.
There was a problem hiding this comment.
Well, we will always subsample every fastq that has more than the max number of reads, right?
2 participants
No description provided.