aha
I'm gonna run a quick svdquartet example in the workshop
it's working well enough
I'm probably going to add dendropy as a dependency to ipyrad.analysis eventually so we can do fancier things with trees. Either that or ete2, but I think dendropy is a bit more lightweight.
sweet, sounds good.
How was the stacks talk?
his workshop is waaaaay more formal than mine
Formal is boring.
I'm just gonna walk them through the docs, run through the tutorials with them, and then set them loose on a test data set.
and tease them a bunch with the new features that aren't ready yet, lol.
That's how you do it. Nobody learns by having some schmo jabber at you all day, you learn by doing.
Lol, nice.
yeah. I'm going to teach them some basic fun stuff too. Like how to use conda and jupyter notebooks
I think the server administrator here is annoyed that I want them all to install their own miniconda
he's adamant about loading a system wide ipyrad, even though I keep telling him that I don't want that, users should have their own sandboxed modules. It doesn't take that much space.
he's trying to install ipyrad and dependencies from source.
and I'm just like, good luck.
lol, glhf
he's seems pro, though. He might get it working.
I'll probably just have them run it locally on 4 cores instead of on the server for simplicity.
btw, is it easy to add 'dendropy' from the channel 'biopython' to our conda.recipe?
or should we host it to our own channel?
probably better to add from their channel so we don't have to deal with versioning, i'll look into it.
ok
I'm finally filling in the parameter explanations to the docs.
been putting it off cuz it's boring.
lol, yeah, boring but important.
refseq mapping for pair data has been fsck for a while, i'm fixing it
yeah, I'm not expecting that to be ready tomorrow. That sounds like a hard problem
I had to make this change to make refseq work on the latest pull (10 seconds ago):
out.write("".join(seqs))
out.write("//\n//\n")
flipped it around so
//\n//\n comes after the seqs
otherwise I get an error
but maybe that is causing a problem...
What does it do if the
//'s are before the seqs?
--------------------------------------------------
ipyrad [v.0.1.73]
Interactive assembly and analysis of RADseq data
--------------------------------------------------
New Assembly: cli
ipyparallel setup: Local connection to 4 Engines
Step1: Demultiplexing fastq data to Samples
Saving Assembly.
Step2: Filtering reads
Saving Assembly.
Step3: Clustering/Mapping reads
Reference sequence index exists
[Engine Exception]
---------------------------------------------------------------------------IndexError Traceback (most recent call last)<string> in <module>()
/home/deren/Documents/ipyrad/ipyrad/assemble/cluster_within.pyc in muscle_align(args)
149 if len(names) <= 1:
150 if names:
--> 151 stack = [names[0]+"\n"+seqs[0]]
152 else:
153 ## split seqs if paired end seqs
IndexError: list index out of range
None
Saving Assembly.
Step4: Joint estimation of error rate and heterozygosity
Saving Assembly.
ipyrad error -
No Samples ready for joint estimation. First run step3().//
//
>1A_0_2813_r1;size=19;*
TGCAGACTGTGACCGGATATGTTGGGGATTTAAAGAGATACGCGCCAAGACGGTGAGGAGGCGAACACCGTTGTACGTCTGGACATTGTTCGGG
>1A_0_2816_r1;size=1;+
TGCAGACTGTGACCGGATATGTTGGGGATTTAAAGAGATACGCGCCAAGTCGGTGAGGAGGCGAACACCGTTGTACGTCTGGACATTGTTCGGG
//
//
>1A_0_6303_r1;size=20;*
TGCAGAAAAAGCGCGGAGAACTATGGTCACACCTAATTCAAACCTCAGCCAAGCATGCGGTTCGGGATCAGTCAAATACTGTGTTCCTAGAATT
>1A_0_6320_r1;size=1;+
TGCAGAAAAAGCGCGGAGAACTATGGTCAGACCTAATTCAAACCTCAGCCAAGCATGCGGTTCGGGATCAGTCAAATACTGTGTTCCTAGAATT
//
//
>1A_0_2197_r1;size=18;*
TGCAGTCTGCGCCCGGGATGGATAACTTCCCTCAGTGCCGTAGACCAATAAAATGTTCACCGTCTGTGCGACAAATGATCGCCAGGCTCACAGC
>1A_0_4543_r1;size=12;+
TGCAGTCCTACCATGAATGGAGTTTCTTCCTAGTTCAACATCCCCGTAATGTAATTGTAGGTGTCGAGCTTCACATGCGACCAGTAAGGAACAT
>1A_0_4547_r1;size=1;+
TGCAGTCCTACCATGAATGGAGTTTCTTCCCAGTTCAACATCCCCGTAATGTAATTGTAGGTGTCGAGCTTCACATGCGACCAGTAAGGAACAT
//
//
that's the top of clust.gz when // comes first. The third cluster down is bad, too.
huh, i guess you're right, must have thought it fixed stuff earlier when i was messing with the messy sim genome, good catch.
i pushed the fix (had a couple other PE fixes in refmap as well).
Going off line for about an hour, i'll check back in in a bit...
i'm back if you need anything
Thanks, I'm out for dinner and drinks so probably won't have much more chance to make changes tonight. Thanks for crunching it today. If you want to add any more to the docs or the -i info that would help.
If you're able to push v.0.1.74 to anaconda for mac in the next hour that would be great. Otherwise I can try to do it here when I get to the mac lab.
done
good luck today! Let me know if there's anything i can do to help.
Awesome
It's working
:clap:
ugh, look at rtd. 10 failed build in a row. The site is down I htink
and it won't load the changes I made last night.
so we're having a tiny hiccup, but things are going pretty well.
I'll check it out
it's a server error, I don't think its our fault
rtfd/readthedocs.org#2002 <- yeh server side :thumbsdown:
The workshop rocked.
that great! psyched it went well. Did you get any good/useful feedback?