I just tried running seqmagick convert --squeeze --include-from-file seq-id-file input.fasta output.fasta, hoping that I'd be able to "squeeze" out all of the columns which end up being all gaps when I select out just the specified set of sequences. Unfortunately, this doesn't work. My guess is that --squeeze only looks at column compositions before downsampling. Admittedly this can be worked around doing a two step pipe thing (seqmagick convert --include-from-file seq-id-file input.fasta - | seqmagick convert --squeeze - output.fasta), but I wonder what the "least surprise" here is, and whether this case isn't worth handling here?
I just tried running
seqmagick convert --squeeze --include-from-file seq-id-file input.fasta output.fasta, hoping that I'd be able to "squeeze" out all of the columns which end up being all gaps when I select out just the specified set of sequences. Unfortunately, this doesn't work. My guess is that--squeezeonly looks at column compositions before downsampling. Admittedly this can be worked around doing a two step pipe thing (seqmagick convert --include-from-file seq-id-file input.fasta - | seqmagick convert --squeeze - output.fasta), but I wonder what the "least surprise" here is, and whether this case isn't worth handling here?