Select features with highest average abundance
.selectTopFeatures(input, top_n)
input
data.table
top_n
number of top features to select