dreycey / phagefilter Goto Github PK

https://github.com/Dreycey/PhageFilter/blob/963095238b15e163024ea98fc598cc89798961a4/src/main.rs#L301C17-L319

I see around here that PhageFilter is manually writing output as fasta format. It is likely that input reads will be fastq format, where the qualities are definitely relevant and should be maintained in the output. I would love to see the format inherited.

Otherwise, it might be simpler to write only IDs and pass the responsibility to read extraction to another tool, such as samtools.

It seems that the format of input reads is determined by file extension. I would encourage you to document this behaviour.

Also, this is brittle and should ideally encompass multiple conventions for extension syntax to be robust. A better "automagic" approach would be to inspect the begging file contents for a match to either header format.

In the end, you might find it simpler to assume a format and give users the option to declare the other.

Separate issue, but I am being lazy

GZIP read files do not appear to be supported? This really awkward with today's sequencer yields.

The header appears to contain references to mapped genomes, but I do not see this explained in the documentation.

Is it possible to supply a ranking/score for each map? This is perhaps an ignorant question, as I do not know what is returned from searching the gSBT.