Found a bug that prevents FluteMLE from running correctly, specifically when running MAView.
When there are genes (~mouse genes) with no human equivalent, MAView tries to replace the NAs in dd by 0s:
dd[is.na(dd)] = 0
but as this normally happens in character columns, we get a class asignment error, stating that it can't assign class='double' values to a class='character' column.
My guess is that R updated assignment class requirements to make the more strict
This is easily solved by changing the line to dd[is.na(dd)] = 0, thus coercing it into character.
Hi,
Found a bug that prevents FluteMLE from running correctly, specifically when running MAView. When there are genes (~mouse genes) with no human equivalent, MAView tries to replace the NAs in dd by 0s:
dd[is.na(dd)] = 0but as this normally happens in character columns, we get a class asignment error, stating that it can't assign class='double' values to a class='character' column.My guess is that R updated assignment class requirements to make the more strict
This is easily solved by changing the line to
dd[is.na(dd)] = 0, thus coercing it into character.Hope this helps, M