These are chat archives for dereneaton/ipyrad

6th
Dec 2017
Jesse Grismer
@J_Grismer_twitter
Dec 06 2017 15:17
@isaacovercast Just a quick question 2 parts. What does ipyrad do differently with data that is identified as pair3rad versus pairddrad, and for which steps does this matter? Thank you!
ClaudiaPaetzold
@ClaudiaPaetzold
Dec 06 2017 16:06
Hi @isaacovercast and @dereneaton.
I was wondering wether you plan to allow different clustering thresholds for within and between sample clustering? I use RADseq to investigate a ca. 10 mya lineage of ca 50 species and I think being able to lower the threshold for the between-sample-clustering would reflect the long diversification time a lot. I suspect keeping the threshold the same, will cause loci to be 'artificially' split and could produce artifacts in subsequent analyses. Thx
Isaac Overcast
@isaacovercast
Dec 06 2017 16:23
@J_Grismer_twitter The only differences are in steps 1 and 2. In step 1 it uses 2x barcodes for demux'ing. In step 2 it uses unique indexes for removing pcr duplicates. After step 2 it's handled the same as peddrad.
@ClaudiaPaetzold We have talked about this before, and it does make sense. Fortunately there is a very easy way to replicate this behavior, as you can simply set the value of the clustering threshold before running step 3 and then change it before running step 6. Hopefully that is good enough for your purposes.
ClaudiaPaetzold
@ClaudiaPaetzold
Dec 06 2017 18:24
@isaacovercast Thank you so much. I will try that right away. Cheers