These are chat archives for dereneaton/ipyrad

10th
Jul 2017
bbarker505
@bbarker505
Jul 10 2017 02:49
Hey all, does anyone know how to run ipyrad on data that have already been demultiplexed and cleaned (from a different program) - i.e. skip steps 1-2? I keep getting an error message of "Error: You must first run step 1 on the assembly: /home/barkerbr/ipyrad_test/euvi_clust90". When I made the parameters file I called it "euvi_clust90", so that's the assembly name. I put my .fastq files in a folder named "/euvi_clust90_edit" within the project folder, because I thought the program was looking for a folder with the assembly name + _edit. Thanks for any help!
Jenny Archibald
@jenarch
Jul 10 2017 13:35
Hi @bbarker505 You need to start from step 1, even if you have already demultiplexed. You won't need param 2, but instead will put the path to your data under param [4] [sorted_fastq_path]. I wouldn't name your input data folder assembly name + _edit because that is a folder that the program creates itself. Those first couple steps should run quickly.
rtm79
@rtm79
Jul 10 2017 17:39
Hi @dereneaton @isaacovercast, tried to install on an hpc, but got the following error: conda install -c ipyrad ipyrad
Fetching package metadata ...
CondaHTTPError: HTTP None None for url https://conda.anaconda.org/ipyrad/linux-64/repodata.json
Elapsed: None
An HTTP error occurred when trying to retrieve this URL.
HTTP errors are often intermittent, and a simple retry will get you on your way.
SSLError(SSLError(SSLError("bad handshake: Error([('SSL routines', 'ssl3_get_server_certificate', 'certificate verify failed')],)",),),)
Any ideas?
Thanks
Deren Eaton
@dereneaton
Jul 10 2017 17:43
@rtm79 my guess is that it was an HTTP error with conda (not a problem with ipyrad) and that retrying in a minute will likely solve the problem.
Otherwise, it looks like a solution others have found for this conda problem (conda/conda#4930) is to set the certificate argument to false for conda: conda config --set ssl_verify False.
rtm79
@rtm79
Jul 10 2017 17:50
Thank you @dereneaton, I kept retrying, but it didn't work. when i tried the latter solution you found ( conda config --set ssl_verify False), the installation proceeded immediately...
Deren Eaton
@dereneaton
Jul 10 2017 17:53
:thumbsup:
Deren Eaton
@dereneaton
Jul 10 2017 22:03

New version 0.7.2 now available on conda (for linux now, for OSX soon)
New features:

  • step 1 now merges technical replicates if identical names in barcodes file.
  • multiple barcodes are checked in step2 filtering if sample is from technical replicates.
  • step 6 CLI now has checkpointing so interrupted analyses can be restarted.
  • step 6 API allows even more fine-scale restarting.
  • bucky analysis tool now available.
  • prettier printing/spacing in the API versus CLI
  • Allow ipcluster profiles in both API and CLI
  • updates to analysis cookbooks for tetrad, bpp, bucky, raxml, treemix.

Bug fixes:

  • pysam conda recipe updated to retrieve newer version
  • bpp analysis tool bug fix for full path names, fixes name conflicts.
  • Fix to allow for subsampling nquartets in tetrad.

  • Please let me know if you run into any problems during upgrades/installation as we made some recent changes to the conda recipe *

@all v.0.7.2 now available :arrow_heading_up: