These are chat archives for dereneaton/ipyrad

May 2016
Deren Eaton
May 26 2016 16:45
@isaacovercast, on my branch I removed the last two params (excludes and outgroups) since they are now deprecated. This breaks backwards compatibility slightly, and only for the CLI. Users will need to create a new params file in the latest ipyrad, and fill in their params. Seems like a good opportunity to tag v.0.3.0. I might change some more params around before doing so, tho. I should have a little time in the next few days to do some work while we're in Mexico City. I have a much improved push for demultiplexing coming as well.
@isaacovercast , clust_threshold is the sequence similarity of the regions which are overlapping. query_cov is how much of the sequence must be overlapping. Specifically, I think, how much of the query sequence must be overlapping with the target (seed). In PE ezrad it is likely that a low overlap (partially overlapping contigs) could cause a problem, tho I thought we had that working pretty well. It also seems reasonable that PE ezrad could have a very high htemp, but that shouldn't cause a crash... @Cycadales, The PE ezrad data should definitely be checked for adapters in step2, since the very common cutters used in that method often lead to adapter contamination.
May 26 2016 23:51
@dereneaton I had the clust_threshold set at 0.85. with regards to how much sequence is overlapping I am not sure that should be a problem because the average BP of my lib
May 26 2016 23:59
opps! sorry I need to finish that sentence. So based on the LabChip read from my final libraries the average BPs size was just over 530 so that should not have too much of a problems with overlap. I did check for adapters at setting one 1 maybe I could do a even more fuzzy check at 2? Can I ask how common cutters would lead to adapter contamination? With each library prep prep I always run a LabChip for about 1/4 of the samples to check for adapter dimer and then do a 0.9 x beads clean. Also the final pooled library goes though QC to check for adapters again. I did get a very minor adapter dimer contamination (very minor) on the first sequencing run but noting on the last two.