Identification and correction of abnormal, incomplete and mispredicted proteins in public databases

Abstract

MisPred works efficiently in identifying errors in predictions generated by the most reliable gene prediction tools such as the EnsEMBL and NCBI's GNOMON pipelines and also guides the correction of errors. We suggest that application of the MisPred approach will significantly improve the quality of gene predictions and the associated databases.