These are chat archives for dereneaton/ipyrad

27th
Jun 2016
danielyao12
@danielyao12
Jun 27 2016 12:48
I have a problem about my data that sequenced by pairddRAD-like technics ( restriction digestion using two different enzymes) and already removed the adapters. Does it mean my data already finish the Step 2 if considering in ipyrad procedure? Specifically, how should I start analyses of my data? Could someone help me with my problem? If possible, I can show my data by email. Thank you very much in advance!
Deren Eaton
@dereneaton
Jun 27 2016 13:57
@danielyao12 You must always start from step1. If your data are already sorted by barcode/index then you should enter the location of your data files in the sorted_fastq_pathparameter; otherwise, enter the location as the raw_fastq_path.
Then run step 2. You must run this step even if you turn off all of the filtering options in ipyrad, because it is necessary to decompress and format the files for step3.
Daren Card
@darencard
Jun 27 2016 18:25
I have one quick question about designating populations and missing data in ipyrad. I have RADseq from 2 populations and want to set my missing data to a max of 50% for each population individually. It appears that the 'min_samples_locus' option is what I want to set the missing data, but it appears this gets applied to all samples from across both populations if I run everything at once. This obviously isn't ideal given my circumstance. Maybe I'm missing something, but is there an easy way to do set missing data per designated population?
Deren Eaton
@dereneaton
Jun 27 2016 20:08
@darencard : We haven't implemented this as a feature yet, but I can bump it up as a priority.
Daren Card
@darencard
Jun 27 2016 20:15
@dereneaton That would be great. In the meantime, maybe I can figure out a workaround. I notice a merge option in the CLI arguments that merges assemblies. Could that work to merge individual assemblies from the 2 populations together? Or is the idea with that option much more limited?
Deren Eaton
@dereneaton
Jun 27 2016 20:21
At the moment it is limited to merging two assemblies created from two separate demultiplexing events.
but implementing the per-population filtering in step 7 should be fairly simple.