dunovo: use low frequency variants, rather than rare variants

The dunovo tutorial provides a great introduction to consensus correction and calling non-diploid low frequency variants:

http://galaxyproject.github.io/training-material/topics/variant-analysis/tutorials/dunovo/tutorial.html

My suggestion would be to refer to the detection consistently as low frequency variant detection, rather than rare variant detection. In my experience with the nomenclature, rare variants refer to variants that are in low abundance in a reference population, like ExAC or 1000 genomes for human samples. When I see rare variant detection, I think you'll be comparing against one of these external database and filtering variants using them with a tool like GEMINI.

For variants that are at low abundance in a sequenced population, referring to them as low frequency variants would be less confusing.

Thanks again for the great work on this tutorial and hope this helps.