Alignment overview

[matsen]

Hey folks-- as I've been poking around with @lauranoges sequences, I've found it useful to look at a zoomed-out view of a multiple sequence alignment of the clonal family sequences before any trimming or pruning. Here's the view of the BF520.1-k-IgH family:

In this case, we see one sequence with a quite surprisingly big insertion in the V, which seems like it might be some sort of preprocessing error? I didn't think that we were getting reads that long. In any case looking at data is a good thing.

Does this seem useful, Laura? Please say if "meh".

If so, it just needs two steps:

• making a multiple sequence alignment of the original sequences
• running seqs2png

The blocks representing the MSA bases can be made bigger and smaller. It's currently 3x3 by default.

[lauradoepker]

Yes that's GREAT! Let's keep it and do for all families, if it's that easy.

[matsen]

OK, do you prefer that view or one with slightly smaller bases:

Note for potential implementors: this was the input_seqs column.

[lauradoepker]

Slightly smaller is fine because we are gaining a birds eye view and nothing else anyway. Sweet.

[matsen]

Although it seems useful to do something with minimally-processed sequences, I don't see how this would fit into our current Olmsted world, so I suggest closing.

[metasoarous]

We could still include this in the core cft pipeline so that the targets are there for folks to poke around at if an issue comes up with a cluster.