Errors in long read assemblies can critically affect protein prediction

Long read, single molecule sequencing technologies are now routinely used for whole-genome sequencing and assembly. However, even after multiple rounds of correction, many errors can remain which can critically affect protein coding regions, resulting in significantly altered and often truncated protein predictions.