Sylvain Poux

Expert curation is essential to capture knowledge of enzyme functions from the scientific literature in FAIR open knowledgebases but cannot keep pace with the rate of new discoveries and new publications. In this work we present EnzChemRED, for Enzyme Chemistry Relation Extraction Dataset, a new training and benchmarking dataset to support the development of Natural Language Processing (NLP) methods such as (large) language models that can assist enzyme curation. EnzChemRED consists of 1,210 expert curated PubMed abstracts where enzymes and the chemical reactions they catalyze are annotated using identifiers from the protein knowledgebase UniProtKB and the chemical ontology ChEBI. We show that fine-tuning language models with EnzChemRED significantly boosts their ability to identify proteins and chemicals in text (86.30% F₁ score) and to extract the chemical conversions (86.66% F₁ score) and the enzymes that catalyze those conversions (83.79% F₁ score). We apply our methods to abstracts at PubMed scale to create a draft map of enzyme functions in literature to guide curation efforts in UniProtKB and the reaction knowledgebase Rhea.

The Gene Ontology (GO) knowledgebase (http://geneontology.org) is a comprehensive resource concerning the functions of genes and gene products (proteins and noncoding RNAs). GO annotations cover genes from organisms across the tree of life as well as viruses, though most gene function knowledge currently derives from experiments carried out in a relatively small number of model organisms. Here, we provide an updated overview of the GO knowledgebase, as well as the efforts of the broad, international consortium of scientists that develops, maintains, and updates the GO knowledgebase. The GO knowledgebase consists of three components: (1) the GO—a computational knowledge structure describing the functional characteristics of genes; (2) GO annotations—evidence-supported statements asserting that a specific gene product has a particular functional characteristic; and (3) GO Causal Activity Models (GO-CAMs)—mechanistic models of molecular “pathways” (GO biological processes) created by linking multiple GO annotations using defined relations. Each of these components is continually expanded, revised, and updated in response to newly published discoveries and receives extensive QA checks, reviews, and user feedback. For each of these components, we provide a description of the current contents, recent developments to keep the knowledgebase up to date with new discoveries, and guidance on how users can best make use of the data that we provide. We conclude with future directions for the project.

The UniProt Knowledgebase UniProtKB is a comprehensive, high-quality, and freely accessible resource of protein sequences and functional annotation that covers genomes and proteomes from tens of thousands of taxa, including a broad range of plants and microorganisms producing natural products of medical, nutritional, and agronomical interest. Here we describe work that enhances the utility of UniProtKB as a support for both the study of natural products and for their discovery. The foundation of this work is an improved representation of natural product metabolism in UniProtKB using Rhea, an expert-curated knowledgebase of biochemical reactions, that is built on the ChEBI (Chemical Entities of Biological Interest) ontology of small molecules. Knowledge of natural products and precursors is captured in ChEBI, enzyme-catalyzed reactions in Rhea, and enzymes in UniProtKB/Swiss-Prot, thereby linking chemical structure data directly to protein knowledge. We provide a practical demonstration of how users can search UniProtKB for protein knowledge relevant to natural products through interactive or programmatic queries using metabolite names and synonyms, chemical identifiers, chemical classes, and chemical structures and show how to federate UniProtKB with other data and knowledge resources and tools using semantic web technologies such as RDF and SPARQL. All UniProtKB data are freely available for download in a broad range of formats for users to further mine or exploit as an annotation source, to enrich other natural product datasets and databases.

Biological processes are accomplished by the coordinated action of gene products. Gene products often participate in multiple processes, and can therefore be annotated to multiple Gene Ontology (GO) terms. Nevertheless, processes that are functionally, temporally and/or spatially distant may have few gene products in common, and co-annotation to unrelated processes probably reflects errors in literature curation, ontology structure or automated annotation pipelines. We have developed an annotation quality control workflow that uses rules based on mutually exclusive processes to detect annotation errors, based on and validated by case studies including the three we present here: fission yeast protein-coding gene annotations over time; annotations for cohesin complex subunits in human and model species; and annotations using a selected set of GO biological process terms in human and five model species. For each case study, we reviewed available GO annotations, identified pairs of biological processes which are unlikely to be correctly co-annotated to the same gene products (e.g. amino acid metabolism and cytokinesis), and traced erroneous annotations to their sources. To date we have generated 107 quality control rules, and corrected 289 manual annotations in eukaryotes and over 52 700 automatically propagated annotations across all taxa.

Motivation: To provide high quality computationally tractable enzyme annotation in UniProtKB using Rhea, a comprehensive expert-curated knowledgebase of biochemical reactions which describes reaction participants using the ChEBI (Chemical Entities of Biological Interest) ontology. Results: We replaced existing textual descriptions of biochemical reactions in UniProtKB with their equivalents from Rhea, which is now the standard for annotation of enzymatic reactions in UniProtKB. We developed improved search and query facilities for the UniProt website, REST API and SPARQL endpoint that leverage the chemical structure data, nomenclature and classification that Rhea and ChEBI provide. Availability and implementation: UniProtKB at https://www.uniprot.org; UniProt REST API at https://www.uniprot.org/help/api; UniProt SPARQL endpoint at https://sparql.uniprot.org/; Rhea at https://www.rhea-db.org.

Motivation: To provide high quality computationally tractable enzyme annotation in UniProtKB using Rhea, a comprehensive expert-curated knowledgebase of biochemical reactions which describes reaction participants using the ontology ChEBI (Chemical Entities of Biological Interest). Results: We replaced existing textual descriptions of biochemical reactions in UniProtKB with their equivalents from Rhea, which is now the standard for annotation of enzymatic reactions in UniProtKB. We developed improved search and query facilities for the UniProt website, REST API, and SPARQL endpoint that leverage the chemical structure data, nomenclature, and classification that Rhea and ChEBI provide. Availability and Implementation: UniProtKB at https://www.uniprot.org/ ; UniProt REST API at https://www.uniprot.org/help/api ; UniProt SPARQL endpoint at https://sparql.uniprot.org/sparql ; Rhea at https://www.rhea-db.org/ . Contact: anne.morgat@sib.swiss

Personalized genomic medicine depends on integrated analyses that combine genetic and phenotypic data from individual patients with reference knowledge of the functional and clinical significance of sequence variants. Sources of this reference knowledge include the ClinVar repository of human genetic variants, a community resource that accepts submissions from external groups, and UniProtKB/Swiss-Prot, an expert-curated resource of protein sequences and functional annotation. UniProtKB/Swiss-Prot provides knowledge on the functional impact and clinical significance of over 30 000 human protein-coding sequence variants, curated from peer-reviewed literature reports. Here we present a pilot study that lays the groundwork for the integration of curated knowledge of protein sequence variation from UniProtKB/Swiss-Prot with ClinVar. We show that existing interpretations of variant pathogenicity in UniProtKB/Swiss-Prot and ClinVar are highly concordant, with 88% of variants that are common to the two resources having interpretations of clinical significance that agree. Re-curation of a subset of UniProtKB/Swiss-Prot variants according to American College of Medical Genetics and Genomics (ACMG) guidelines using ClinGen tools further increases this level of agreement, mainly due to the reclassification of supposedly pathogenic variants as benign, based on newly available population frequency data. We have now incorporated ACMG guidelines and ClinGen tools into the UniProt Knowledgebase (UniProtKB) curation workflow and routinely submit variant data from UniProtKB/Swiss-Prot to ClinVar. These efforts will increase the usability and utilization of UniProtKB variant data and will facilitate the continuing (re-)evaluation of clinical variant interpretations as data sets and knowledge evolve.

High-throughput studies constitute an essential and valued source of information for researchers. However, high-throughput experimental workflows are often complex, with multiple data sets that may contain large numbers of false positives. The representation of high-throughput data in the Gene Ontology (GO) therefore presents a challenging annotation problem, when the overarching goal of GO curation is to provide the most precise view of a gene's role in biology. To address this, representatives from annotation teams within the GO Consortium reviewed high-throughput data annotation practices. We present an annotation framework for high-throughput studies that will facilitate good standards in GO curation and, through the use of new high-throughput evidence codes, increase the visibility of these annotations to the research community.

Biological knowledgebases, such as UniProtKB/Swiss-Prot, constitute an essential component of daily scientific research by offering distilled, summarized and computable knowledge extracted from the literature by expert curators. While knowledgebases play an increasingly important role in the scientific community, their ability to keep up with the growth of biomedical literature is under scrutiny. Using UniProtKB/Swiss-Prot as a case study, we address this concern via multiple literature triage approaches. With the assistance of the PubTator text-mining tool, we tagged more than 10 000 articles to assess the ratio of papers relevant for curation. We first show that curators read and evaluate many more papers than they curate, and that measuring the number of curated publications is insufficient to provide a complete picture as demonstrated by the fact that 8000–10 000 papers are curated in UniProt each year while curators evaluate 50 000–70 000 papers per year. We show that 90% of the papers in PubMed are out of the scope of UniProt, that a maximum of 2–3% of the papers indexed in PubMed each year are relevant for UniProt curation, and that, despite appearances, expert curation in UniProt is scalable. UniProt is freely available at http://www.uniprot.org/. Supplementary data are available at Bioinformatics online.

Bacterial viruses, also called bacteriophages, display a great genetic diversity and utilize unique processes for infecting and reproducing within a host cell. All these processes were investigated and indexed in the ViralZone knowledge base. To facilitate standardizing data, a simple ontology of viral life-cycle terms was developed to provide a common vocabulary for annotating data sets. New terminology was developed to address unique viral replication cycle processes, and existing terminology was modified and adapted. Classically, the viral life-cycle is described by schematic pictures. Using this ontology, it can be represented by a combination of successive events: entry, latency, transcription/replication, host–virus interactions and virus release. Each of these parts is broken down into discrete steps. For example enterobacteria phage lambda entry is broken down in: viral attachment to host adhesion receptor, viral attachment to host entry receptor, viral genome ejection and viral genome circularization. To demonstrate the utility of a standard ontology for virus biology, this work was completed by annotating virus data in the ViralZone, UniProtKB and Gene Ontology databases.