What this means is that civic data contain various types of biomarker types, including things like 'overexpression' (which is not actually a problem, as this just means increased levels of ) and mutations and maybe others as well. Also, there is no cross-reference to the actual entity identifier in column D of the output. Instead the gene is indicated there. The gene information is useful, but is not the correct thing to put in that column. I'd expect something like a dbSNP identifier.
mixed bag of things. will look into it and discuss with Darren on what needs to be done