Merge branch 'master' of https://github.com/dereneaton/ipyrad

docs/faq.rst

@@ -318,3 +318,12 @@ While it does seem that the # of pis shouldn't change under varying `max_snp_loc
 Can ipyrad assemble MIG-seq data?
 ---------------------------------
 MIG-seq (multiplexed ISSR genotyping by sequencing) is a method proposed by Suyama and Matsuki (2015), which involves targeting variable regions between simple sequence repeats (SSR). The method produces data that is somewhat analogous to ddRAD, in that you have the variable region which is flanked on either side by sequences that are known to be repeated randomly and at some appreciable frequency throughout the genome. Check out the `figure from the manuscript <https://www.nature.com/articles/srep16963/figures/1>`__. Anyway.... yes, ipyrad can assemble this kind of data, though there are some tricks. Primarily we recommend higher values of `filter_min_trim_len` and `clust_threshold`. If sequenced on a desktop NGS platform (Ion Torrent PGM, MiSeq) it also helps to reduce both `mindepth` params to recover more clusters.
+
+Why are my ipcluster engines dying silently on cluster compute notes?
+---------------------------------------------------------------------
+This is a nasty bug that's bitten me more than once. If you are having trouble with cluster engines running jobs and then dying silently it may be because the cluster is headless and the engines are trying to interact with a GUI backend. This causes nasty things to happen. Here are a couple links that provide workable solutions:
+
+https://groups.google.com/a/continuum.io/forum/#!topic/anaconda/o0pnE9PEqA0
+
+https://github.com/ipython/ipyparallel/issues/213
+