fasta --disambiguate feature request

[mmelendrez]

Can we build an option to 'force' permutations? I know we set an upper boundary of 100 which I still think is good - but when recreating a 'quasispecies' and it tells me it'll create 128 permutations of the sequence because it has 7 ambiguities - I kind of do what to recreate them to look at the theoretical population.

[averagehat]

We could do either/both of:
a) let the user set the upper bound themselves (this would require re-running the script in cases like this)
b) prompt the user with [Y/n] to see if they want to create all permutations (this might be annoying if you are doing a lot of reads with many permutations)

[mmelendrez]

I like both options - I'd probably use the prompt version but I get that this would be annoying if there were many sequences you had to do this with. Could create a flag --prompt which, when added to the script, turns prompting on? Otherwise the PI could go and change --upper bound [digit]. You could also build in a memory warning with a continue [Y/n]?

[mmelendrez]

Ah additionally - I can't remember if we'd talked about this - but 'if' in the alignment that is given to the script there is a sequence that doesn't have ambiguities (because that just happened to me so I know it'll happen with other investigators) it should skip the sequence (obviously) BUT still add it to the final fasta output.

[necrolyte2]

I like the idea of default is to just prompt every time it goes over certain amount or you can increase the limit

--no-prompt
--max-permutations

[averagehat]

@mmelendrez does it currently not output the fasta sequence if it isn't ambiguous?

[necrolyte2]

I thought it would, but now curious

[mmelendrez]

It did not in this last run I did

[mmelendrez]

I only got the sequences back that had ambiguities. It spit back out the one it could not disambiguate but I had another sequence in there that had no ambiguities...here let me find the fasta and I'll attach

[mmelendrez]

ZikaThai_MinorVar.fas.txt