Annotation Error in Public Databases: Misannotation of Molecular Function in Enzyme Superfamilies

By Alexandra M. Schnoes, Shoshana D. Brown, Igor Dodevski, Patricia C. Babbitt. 2009

736 citations recorded by the source corpus.

Due to the rapid release of new data from genome sequencing projects, the majority of protein sequences in public databases have not been experimentally characterized; rather, sequences are annotated using computational analysis. The level of misannotation and the types of misannotation in large public databases are currently unknown and have not been analyzed in depth. We have investigated the misannotation levels for molecular function in four public protein sequence databases (UniProtKB/Swiss-Prot, GenBank NR, UniProtKB/TrEMBL, and KEGG) for a model set of 37 enzyme families for which extensive experimental information is available. The manually curated database Swiss-Prot shows the lowest annotation error levels (close to 0% for most families); the two other protein sequence databases (GenBank NR and TrEMBL) and the protein sequences in the KEGG pathways database exhibit similar and surprisingly high levels of misannotation that average 5%–63% across the six superfamilies studied. For 10 of the 37 families examined, the level of misannotation in one or more of these databases is >80%. Examination of the NR database over time shows that misannotation has increased from 1993 to 2005. The types of misannotation that were found fall into several categories, most associated with “overprediction” of molecular function. These results suggest that misannotation in enzyme superfamilies containing multiple families that catalyze different reactions is a larger problem than has been recognized. Strategies are suggested for addressing some of the systematic problems contributing to these high levels of misannotation.

Subfields: Molecular Biology, Spectroscopy

How researchers see this paper

The researchers whose trusted networks rank it highest, strongest advocate first. Every researcher gets a different order.

In Molecular Biology

In Spectroscopy

Citations in the corpus

NetSurfP-2.0: improved prediction of protein structural features by integrated deep learning (2018, background): “tions is a well-known and unresolved problem affecting public databases.(8) Such errors often propagate through databases and sequence”
Enzyme Function Initiative-Enzyme Similarity Tool (EFI-EST): A web tool for generating protein sequence similarity networks (2015, positive): “However, a conservative estimate is that only ~50% of the proteins discovered in genome projects have reliable functional annotations in the sequence databases—the remainder have unknown, uncertain, or incorrect functional annotations (and the identities of these are unknown!) [1, 2].”
Ortholog identification in the presence of domain architecture rearrangement (2011, neutral): “notated [5].”
Embracing the unknown: disentangling the complexities of the soil microbiome (2017, neutral): “Third, there are methodologi cal issues associated with linking specific taxa to specific metabolic processes; these include problems with accurate gene annotation 76 , insufficient taxonomic resolution, and the rapid turnover of transcripts, proteins and metabolites 77”
TIGRFAMs and Genome Properties in 2013 (2012, background): “In fact, misannotations that attach overly speciﬁc functions are a particularly abundant class of error (16).”
The SIB Swiss Institute of Bioinformatics’ resources: focus on curated databases (2015, background): “The quality and utility of the information provided is far superior to that which is achieved by using computational means (15).”
HAMAP in 2015: updates to the protein family classification and annotation system (2014, methodology): “Manual curation of the experimental literature in UniProtKB/Swiss-Prot is highly accurate (6), with expert curation of HAMAP profiles and rules specifically designed to avoid over-annotation through the careful definition of isofunctional protein families and functionally important residues.”
An In silico approach for the evaluation of DNA barcodes (2010, background): “Furthermore, annotation errors are present in Genbank [25], and the error rate might be clade dependent.”
Detection of phytopathogens of the genus Dickeya using a PCR primer prediction pipeline for draft bacterial genome sequences (2013, methodology): “More broadly, it may be appropriate, in the context of the rapidly increasing numbers of bacterial genome sequences being generated, to assign annotation quality or evidence codes to taxonomic classification in sequence databases in a similar fashion to those used and proposed for protein functional”
Expert curation in UniProtKB: a case study on dealing with conflicting and erroneous data (2014, background): “…molecular function in four public protein sequence databases for a set of 37 enzyme families for which extensive experimental information is available and concluded that the level of erroneous annotation was much higher in automatically annotated databases than in manually curated resources (2).”
Large-Scale Determination of Sequence, Structure, and Function Relationships in Cytosolic Glutathione Transferases across the Biosphere (2014, background): “Swiss-Prot is a manually curated database of protein sequences shown to have among the most reliable annotations that can be obtained from large online databases [34].”