Cannot read sequences containing '-' gaps

[blake-riley]

If the input FASTA contains a '-' gap, then the FASTA file is rejected for containing illegal sequence characters.

Should QuickProbs be able to realign "gappy" sequences?

As a minimum working example, if I have:
in.fasta =

>seq2
ALIG-N

then the following command will fail:

~/Downloads/quickprobs-2.06-osx -v in.fasta

with the following message:

Loading sequence file: in.fasta...illegal sequence character:-
Illegal characters in sequence set!

--
OS: macOS 10.12.6
ver: release quickprobs-2.06-osx

[agudys]

Hello,
Unfortunately current version of QuickProbs does not allow realignment of sequences with gaps. The only way around is remove gaps and rerun the alignment, but this, obviously won't take into account gap information in the model. You may also try to realign only regions of interest. If solution with gap removal is convinient for you, we can built it into software.

[blake-riley]

Hi Adam!
It wasn't clear from the user manual / FAQs that QuickProbs couldn't accept sequences with gaps.

To clarify my use case: I've been calling sequence alignment software (including QuickProbs) from within a sequence alignment viewer, to realign sequences and regions of interest.
Currently, I'm using a script that calls sed to strip '-' characters from an input fasta, before passing the fasta off to QuickProbs. This works fine for me, and I am happy with this workaround.

I imagine that QuickProbs might be a little more accessible to others, if it removed the gaps automatically before running its alignment?

[agudys]

We're gonna fix this in the next release. Let me know if you have some other issues or suggestions!