Add verification that sequence looks like sequence

[DavidEHufnagel]

Inside fasta_manager, when a FastaEntry is created it should first check that the sequence looks like sequence, meaning that it is in one of these categories:

1. Is a standard nucleotide base: 'A','T','C','G'
2. Is a nonstandard nucleotide base representing multiple possibilities like 'Y'
3. Represents a lack of data properly with a '-' or an 'N'

[sammysheep]

This functionality does exist in Zoe as well, but no harm in doing it yourself.

[DavidEHufnagel]

I thought Zoe would be good at this. We discussed that Zoe is not public, which is when I considered implementing it personally. That's a barrier to using Zoe. I'm less concerned about having to adopt nightly. If you have a timeline for Zoe I'd like to hear it either in this venue or elsewhere. I'm sure it's also generally worthwhile to consider whether adding a dependency could work well to save development time.

[DavidEHufnagel]

Add to the list here to ensure that the sequence is not empty