search

matsengrp / cft

Clonal family tree
5 stars 3 forks source link

Fix DNA seq color annotations in cftweb #127

Closed metasoarous closed 6 years ago

metasoarous commented 7 years ago

Taking care of #8 and #14 in #126, there were some issues with index errors when we switched to the seqs with indels. This seems to have resolved itself, but we should still check in on the annotations so no one gets tripped up, fix it if it's a problem, and close this if it's not. Would @WSDeWitt or @lauranoges be able to check visually?

lauradoepker commented 7 years ago

@metasoarous I'm looking at stoat:5051. I click on the nucleotide alignment at the bottom and I see the same color scheme (green, blue, orange, purple) as before, along with bolded nucleotides. I'm guessing that green = V, blue = NTI, orange = D, and purple = J genes. I'm not sure what the bold means... perhaps CDR3 region.

This looks the same as it did before, so that in itself may answer your question.

For us, this color scheme, though helpful, would be greatly improved by having mutations different from "sequence of interest" show as red-colored or something.

metasoarous commented 7 years ago

OK; Do the boundaries of the regions highlighted make sense to you? In particular look for an example where you see some insertions or deletions, since that's where we'd be potentially getting mismatch.

We will definitely be annotating and styling the mutations, as per #111. Just haven't gotten around to that yet.

metasoarous commented 7 years ago

I confirmed with @psathyrella that the existing indexing is based on the indel reversed sequences. For now, I'm just taking the annotations out (see fba8f8), but once we get this fixed upstream we can add back in.

metasoarous commented 6 years ago

Removing this since in olmsted we don't even show dna alignment.