Quality Matters: Biocuration Experts on the Impact of Duplication and Other Data Quality Issues in Biological Databases
Loading...
Author/Creator ORCID
Date
2020-04
Type of Work
Department
Program
Citation of Original Publication
Chen, Qingyu et al. Quality Matters: Biocuration Experts on the Impact of Duplication and Other Data Quality Issues in Biological Databases. Genomics, Proteomics & Bioinformatics 18 (April 2020) no. 2, pages 91-103.
Rights
This item is likely protected under Title 17 of the U.S. Copyright Law. Unless on a Creative Commons license, for uses protected by Copyright Law, contact the copyright holder or the author.
Attribution-NonCommercial-NoDerivatives 4.0 International (CC BY-NC-ND 4.0)
Attribution-NonCommercial-NoDerivatives 4.0 International (CC BY-NC-ND 4.0)
Subjects
Abstract
Biological databases represent an extraordinary collective volume of work. Diligently built up over decades and comprising many millions of contributions from the biomedical research community, biological databases provide worldwide access to a massive number of records (also known as entries) [1]. Starting from individual laboratories, genomes are sequenced, assembled, annotated, and ultimately submitted to primary nucleotide databases such as GenBank [2], European Nucleotide Archive (ENA) [3], and DNA Data Bank of Japan (DDBJ) [4] (collectively known as the International Nucleotide Sequence Database Collaboration, INSDC). Protein records, which are the translations of these nucleotide records, are deposited into central protein databases such as the UniProt KnowledgeBase (UniProtKB) [5] and the Protein Data Bank (PDB) [6]. Sequence records are further accumulated into different databases for more specialized purposes: RFam [7] and PFam [8] for RNA and protein families, respectively; DictyBase [9] and PomBase [10] for model organisms; as well as ArrayExpress [11] and Gene Expression Omnibus (GEO) [12] for gene expression profiles. These databases are selected as examples; the list is not intended to be exhaustive. However, they are representative of biological databases that have been named in the “golden set” of the 24th Nucleic Acids Research database issue (in 2016). The introduction of that issue highlights the databases that “consistently served as authoritative, comprehensive, and convenient data resources widely used by the entire community and offer some lessons on what makes a successful database” [13]. In addition, the associated information about sequences is also propagated into non-sequence databases, such as PubMed (https://www.ncbi.nlm.nih.gov/pubmed/) for scientific literature or Gene Ontology (GO) [14] for function annotations. These databases in turn benefit individual studies, many of which use these publicly available records as the basis for their own research.
Inevitably, given the scale of these databases, some submitted records are redundant [15], inconsistent [16], inaccurate [17], incomplete [18], or outdated [19]. Such quality issues can be addressed by manual curation, with the support of automatic tools, and by processes such as reporting of the issues by contributors detecting mistakes. Biocuration plays a vital role in biological database curation [20]. It de-duplicates database records [21], resolves inconsistencies [22], fixes errors [17], and resolves incomplete and outdated annotations [23]. Such curated records are typically of high quality and represent the latest scientific and medical knowledge. However, the volume of data prohibits exhaustive curation, and some records with quality issues remain undetected.
In our previous studies, we (Chen, Verspoor, and Zobel) explored a particular form of quality issue, which we characterized as duplication [24], [25]. As described in these studies, duplicates are characterized in different ways in different contexts, but they can be broadly categorized as redundancies or inconsistencies. The perception of a pair of records as duplicates depends on the task. As we wrote in a previous study, “a pragmatic definition for duplication is that a pair of records A and B are duplicates if the presence of A means that B is not required, that is, B is redundant in the context of a specific task or is superseded by A.” [24]. Many such duplicates have been identified through curation, but the prevalence of undetected duplicates remains unknown, as is the accuracy and sensitivity of automated tools for duplicate or redundancy detection. Other studies have explored the detection of duplicates but often under assumptions that limit the impact. For example, some researchers have assumed that similarity of genetic sequence is the sole indicator of redundancy, whereas in practice, some highly similar sequences may represent distinct information and some rather different sequences may in fact represent duplicates [26]. The notion and impacts of duplication are detailed in the next section.
In this study, the primary focus is to explore the characteristics, impacts, and solutions to duplication in biological databases; and the secondary focus is to further investigate other quality issues. We present and consolidate the opinions of more than 20 experts and practitioners on the topic of duplication and other data quality issues via a questionnaire-based survey. To address different quality issues, we introduce biocuration as a key mechanism for ensuring the quality of biological databases. To our knowledge, there is no one-size-fits-all solution even to a single quality issue [27]. We thus explain the complete UniProtKB/Swiss-Prot curation process, via a descriptive report and an interview with its curation team leader, which provides a reference solution to different quality issues. Overall, the observations on duplication and other data quality issues highlight the significance of biocuration in data resources, but a broader community effort is needed to provide adequate support to facilitate thorough biocuration.