changeset 30:4188853b940b

Update to Miller Lab devshed revision eb4e61d024db
author Richard Burhans <burhans@bx.psu.edu>
date Fri, 26 Jul 2013 12:51:13 -0400
parents fb944979bf35
children a631c2f6d913
files dpmix.xml
diffstat 1 files changed, 1 insertions(+), 1 deletions(-) [+]
line wrap: on
line diff
--- a/dpmix.xml	Thu Jul 25 12:01:47 2013 -0400
+++ b/dpmix.xml	Fri Jul 26 12:51:13 2013 -0400
@@ -73,7 +73,7 @@
 
     <param name="p_input" type="data" format="gd_indivs" label="Potentially admixed individuals" />
 
-    <param name="switch_penalty" type="float" min="0" value="10" label="Genotype switch penalty" help="Note: Depends on the density of SNPs.  For instance, with 50,000 SNPs in a vertebrate genome, 1.0 might be appropriate, with millions of SNPs, a value between 10 and 100 might be reasonable."/>
+    <param name="switch_penalty" type="float" min="0" value="10" label="Genotype switch penalty" help="Note:  The best choice for the Genotype switch penalty depends on the density of SNPs and the age of the admixture events.  With 50,000 SNPs in a vertebrate genome, 10.0 might be appropriate, with millions of SNPs, 100.0 might work better.  We recommend experimenting with various thresholds on minimal spacing between SNVs (to increase independence), minimal FST between the source populations (to identify &quot;ancestry informative markers&quot;), and Genotype switch penalty, to reach conclusions that are robust to changes in analysis parameters."/>
 
     <conditional name="user_het">
       <param name="choice" type="select" format="integer" label="Heterochromatin info">