These are chat archives for dereneaton/ipyrad

22nd
May 2016
Deren Eaton
@dereneaton
May 22 2016 13:38
I have had zero time to work on anything while in Mex, and it looks like it will mostly remain that way for the next two weeks. You making any progress on testing? Any news from the folks doing empirical tests?
roneytan
@roneytan
May 22 2016 16:10
@dereneaton Hi Deren. I did run into a bug, but I was going to wait until you returned before reporting it. I'm running a larger dataset (44 samples dd PE) and everything was working OK, except I was losing a lot of reads due to quality score. I tried to dial down the "phred_Qscore_offset" to 26. Step two showed that I was retaining a lot more reads. However, when I ran step three I received this error:

Step3: Clustering/Mapping reads
[####################] 100% dereplicating | 0:05:41
[####################] 100% clustering | 6:17:10
[####################] 100% chunking | 0:00:10
[####################] 100% aligning | 1:07:27
[####################] 100% concatenating | 0:11:13
Caught unknown exception - sequence item 0: expected string, Sample found
Saving Assembly.

Traceback (most recent call last):
File "/Users/roneytan/miniconda3/envs/pyrad/bin/ipyrad", line 9, in <module>
load_entry_point('ipyrad==0.2.8', 'console_scripts', 'ipyrad')()
File "/Users/roneytan/miniconda3/envs/pyrad/lib/python2.7/site-packages/ipyrad/main.py", line 364, in main
data.run(steps=steps, force=args.force, preview=args.preview)
File "/Users/roneytan/miniconda3/envs/pyrad/lib/python2.7/site-packages/ipyrad/core/assembly.py", line 1376, in run
self.step3(force=force, preview=preview)
File "/Users/roneytan/miniconda3/envs/pyrad/lib/python2.7/site-packages/ipyrad/core/assembly.py", line 1267, in step3
[samples, noreverse, force, preview], 45)
File "/Users/roneytan/miniconda3/envs/pyrad/lib/python2.7/site-packages/ipyrad/core/assembly.py", line 837, in _clientwrapper
stepfunc(args)
File "/Users/roneytan/miniconda3/envs/pyrad/lib/python2.7/site-packages/ipyrad/core/assembly.py", line 1040, in _step3func
force, preview, ipyclient)
File "/Users/roneytan/miniconda3/envs/pyrad/lib/python2.7/site-packages/ipyrad/assemble/cluster_within.py", line 1177, in run
apply_jobs(
args)
File "/Users/roneytan/miniconda3/envs/pyrad/lib/python2.7/site-packages/ipyrad/assemble/cluster_within.py", line 702, in apply_jobs
failed_samples = check_results(async_results)
File "/Users/roneytan/miniconda3/envs/pyrad/lib/python2.7/site-packages/ipyrad/assemble/cluster_within.py", line 780, in check_results
print(" Samples failed this step:"+" ".join(errors.keys()))
TypeError: sequence item 0: expected string, Sample found

After "concatenating" is done, it hangs for a while and then spits out this error. The ipyrad log gives reads:
2016-05-21 04:33:08,133 pid=18208 [assembly.py] ERROR sequence item 0: expected string, Sample found
So, maybe it's because I'm reducing the offset? I can try playing around with it on reduced datasets to see if it might be a sample size issue.
Isaac Overcast
@isaacovercast
May 22 2016 16:20
@roneytan I wouldn't expect changing the phredq score to cause this error, this looks more like a problem with the code, not the data. I assume you're using pairddrad as the assembly method. What version of ipyrad are you running?
I looked at the code and I remember fixing this bug, i'll bet if you pull down the latest version it'll work.
Isaac Overcast
@isaacovercast
May 22 2016 16:25
@dereneaton Yeah, alls good, i'm cranking through glenn's gbs data and have not run into any issues at all so far. All steps running to completion, and all are much much faster than the last time I ran them, so that's a bonus. Hope you're having a good time in mex
roneytan
@roneytan
May 22 2016 16:26
@isaacovercast Hi Isaac. I'm using pairddrad denovo assembly. I pulled down v.2.8.0-1
Isaac Overcast
@isaacovercast
May 22 2016 16:28
Yeah give that a shot, i'll bet it'll fix the problem. Let me know how it goes...
roneytan
@roneytan
May 22 2016 16:31
@isaacovercast My bad, Isaac. I should have told you: I am getting the error with v.2.8.0-1. It's weird because I'm not getting the error unless I change the offset
Isaac Overcast
@isaacovercast
May 22 2016 17:47

Hey ron. Are you sure this is the version that's actually running? The line that is throwing the error doesn't exist anymore in the current codebase:

print(" Samples failed this step:"+" ".join(errors.keys()))

Are you pulling from conda? Can you do an ipyrad -v and let me know the output?

Maybe the conda package is out of date? Are you on mac or linux?
Isaac Overcast
@isaacovercast
May 22 2016 17:52
Either way it does look like changing the phredq is causing samples to fail clustering, which seems weird. I'll have to think about why that would happen. If you dropbox me a couple of the demultiplexed sample fastqs and your params file i can take a look at it.
roneytan
@roneytan
May 22 2016 22:07
@isaacovercast Hi Isaac, i think I pulled v. 0.2.8.0-2 from conda:
ipyrad -v gives me "ipyrad 0.2.8"
sorry 0.2.8.0-1
I'm running mac Yosemite 10.10.5
I'll dropbox some files to you.