Find your elected officials here and spread the word! We're in this together.
secure.ucs.org/a/science-ne...

Finally, to visualize an interaction in this setup, we built an R shiny app where you can compare allele frequencies and interpret the resulting statistics. Give it a try and let me know what you think! shiny.bio.nyu.edu/ms4131/epist...

One of the challenges of this method was building the statistical test to incorporate both genetic background and selection. I wrote an R package to analyze this type of data, and any other comparison of variants by category. github.com/cbuzby/cybrBSA

What if instead we fixed a *region* of the genome, a single chromosome, instead of each locus one by one? We fixed chromosomes I and VIII in budding yeast and then used bulk segregant analysis for the trait of copper resistance to identify QTL - and found interactions!

What if the context in question is another allele? These genetic background effects, or epistasis, can be identified by fixing a locus and searching for all other differentially effective loci - but multiple hypothesis testing reduces statistical power for those results. (3/6)

Let’s say one version of a gene works really well in the sun, but another works really well in the shade. The average of this gene effect might not tell you this variation, and so separating out contexts would provide important information. (2/6)