Can we change the dots/key for the seed sequences on the trees?

matsengrp / cft

Clonal family tree

5 stars 3 forks source link

Can we change the dots/key for the seed sequences on the trees? #163

Closed lauradoepker closed 7 years ago

lauradoepker commented 7 years ago

@metasoarous

Hi there - Megan again from Laura's account.

Can the dot corresponding to the seed be changed to either the same color as the seed lineage dots (maroon/red) OR a completely different color? As it stands, the seeds' labels have been grouped with the other timepoints' labels which isn't as accurate since the seeds originated from a 914dpi sample. (i.e. on QA255.006VL trees the seed's dot is yellow which according the the key is from 462dpi rather than its true timepoint of 914dpi.)

Thanks! Megan

metasoarous commented 7 years ago

So... this is something I've been thinking a little bit about. I assumed that the seed belonged to one sample or the other, but the way things come out in the data it looks like it came from both, and we end up just picking one. So either:

We make sure this information gets passed through from the get go in our input metadata, so we can color with the right timepoint
We don't worry about all that noise and just color the same as the lineage dots as you suggest

2 is a lot easier, but 1 is probably a good idea anyway (for sanity).

I think let's do 2 for now, but in the course of reworking some of the dataflow from partis to cft, we can maybe look at cleaning this up.

lauradoepker commented 7 years ago

Yes, to revisit the info on the seed itself: it came from a biological sort which was from an intermediate date between the two deep sequencing timepoints. It is in it's own 'time' category.

I think the best solution is to have the seeds be their own distinct color (not yellow or orange). I agree with Megan, they might as well be red because that's the color of the seed lineage anyway.

@metasoarous would you like the metadata on all the seeds? I can tell you the timepoints from which they came for the kate-QRS and laura-MB datasets. Let me know

metasoarous commented 7 years ago

Okay... intermediate timepoint corresponds with what I recall from my conversation with @psathyrella.

It's agreed; let's do the separate seedlineage color.

No rush on the seed metadata, but whenever you get around to it.

lauradoepker commented 7 years ago

@metasoarous Here is the link to the metadata that we created a while ago. I've added timepoints to the seeds on the second sheet.

lauradoepker commented 7 years ago

And here's the excel file for the laura-MB metadata so far (only for patient BF520). laura-mb-2016-12-22_library_metadata.xlsx