On the other hand, the potential subtlety of mutations in ncRNAs, even when they are targeted by reverse genetics, is illustrated by the case of BC1, which is expressed in synaptodendritic domains of neurons in rodents. Knockout of this transcript produces no obvious physical or neurological abnormality, and the mutant mice were initially indistinguishable from wild type, but were subsequently found to have reduced exploratory behaviour and consequently a higher mortality in field experiments [100]. Thus this ncRNA causes a subtle behavioural phenotype that is invisible in the cage, and would escape detection in superficial forward genetic screens, but is almost certainly strongly disadvantageous in the wild. Similarly, deletions or insertions in some ultraconserved enhancers yield no discernable abnormality [101],[102], despite the fact that these sequences are clearly under intense selection [103], suggesting insensitivity of phenotypic screens in captivity or redundancy in the regulatory architecture (perhaps associated with developmental robustness) that we have yet to understand.
