eggNOG 4.5: a hierarchical orthology framework with improved functional annotations for eukaryotic, prokaryotic and viral sequences.

Huerta-Cepas J, Szklarczyk D, Forslund K, Cook H, Heller D, Walter MC, Rattei T, Mende DR, Sunagawa S, Kuhn M, Jensen LJ, von Mering C, Bork P. · 2016

Read eggNOG 4.5: a hierarchical orthology framework with improved functional annotations for eukaryotic, prokaryotic and viral sequences. by Huerta-Cepas J, Szklarczyk D, Forslund K, Cook H, Heller D, Walter MC, Rattei T, Mende DR, Sunagawa S, Kuhn M, Jensen LJ, von Mering C, Bork P. (2016) in the Tower of Babel, a free online library of public-domain works preserved in clean, searchable text. No sign-up. The full text is served as a plain-text file at a predictable URL, where text is present: https://antarctic-labs.com/tower-of-babel/europmc-pmc4702882.txt.

Introduction

Jaime Huerta-Cepas, Damian Szklarczyk, Kristoffer Forslund, and nine colleagues released eggNOG 4.5 in Nucleic Acids Research in 2016. eggNOG groups proteins into orthologous groups, sets of genes descended from single ancestral genes in different species. Such groups let researchers transfer functional knowledge between organisms. Version 4.5 reworked how groups nest across taxonomic levels, making them consistent from broad to narrow.

Better nesting let function terms flow into nearly ninety-six thousand previously bare groups, lifting coverage from sixty-seven to seventy-two percent. Each group now carries ontology terms, pathway links, and domain notes. The update adds direct pair-level orthology calls from family trees, a fast mapper for new sequences, and a viral set covering thousands of proteins. A redesigned website speeds searching and browsing. The resource remains free for bulk download, giving evolution studies a current, layered map of gene kinship.

Catalog record

CREATORHuerta-Cepas J, Szklarczyk D, Forslund K, Cook H, Heller D, Walter MC, Rattei T, Mende DR, Sunagawa S, Kuhn M, Jensen LJ, von Mering C, Bork P.
YEAR2016
COLLECTIONPAPERS
FORMATTXT
FILE SIZE36 KB
LICENSECC BY 4.0
RIGHTSCC BY
SOURCEEurope PMC

Source: https://europepmc.org/articles/PMC4702882

Browse the full library · About the Tower of Babel

Full text

The opening of the work, from the library's hosted text:

eggNOG 4.5: a hierarchical orthology framework with improved functional annotations for eukaryotic, prokaryotic and viral sequences By Jaime Huerta-Cepas, Damian Szklarczyk, Kristoffer Forslund, Helen Cook, Davide Heller, Mathias C. Walter, Thomas Rattei, Daniel R. Mende, Shinichi Sunagawa, Michael Kuhn, Lars Juhl Jensen, Christian von Mering Published in: Nucleic Acids Research DOI: 10.1093/nar/gkv1248 ABSTRACT eggNOG is a public resource that provides Orthologous Groups (OGs) of proteins at different taxonomic levels, each with integrated and summarized functional annotations. Developments since the latest public release include changes to the algorithm for creating OGs across taxonomic levels, making nested groups hierarchically consistent. This allows for a better propagation of functional terms across nested OGs and led to the novel annotation of 95 890 previously uncharacterized OGs, increasing overall annotation coverage from 67% to 72%. The functional annotations of OGs have been expanded to also provide Gene Ontology terms, KEGG pathways and SMART/Pfam domains for each group. Moreover, eggNOG now provides pairwise orthology relationships within OGs based on analysis of phylogenetic trees. We have also incorporated a framework for quickly mapping novel sequences to OGs based on precomputed HMM profiles. Finally, eggNOG version 4.5 incorporates a novel data set spanning 2605 viral OGs, covering 5228 proteins from 352 viral proteomes. All data are accessible for bulk downloading, as a web-service, and through a completely redesigned web interface. The new access points provide faster searches and a number of new browsing and visualization capabilities, facilitating the needs of both experts and less experienced users. eggNOG v4.5 is available at http://eggnog.embl.de. INTRODUCTION Orthology and paralogy are central concepts in evolutionary biology. They allow distinguishing between molecular sequences that, despite sharing a common ancestry, evolved by different mechanisms: orthologs are the result of speciation events, whereas paralogs originate from gene duplications. This distinction is widely used in molecular biology, since the evolutionary forces shaping the respective classes of sequences are profoundly different and impact the analysis of functional divergence (1). It is generally assumed that orthologous genes are more likely to conserve their function than paralogs, which, in contrast to orthologs, are partially released from selective pressures after duplication. This idea is commonly referred as the Ortholog Conjecture and, although recently questioned (2,3), it is still considered generally valid and represents the basis of most functional annotation methods (4). Consequently, precise orthology assignments are crucial in many fields such as phylogenetics, pharmacology and comparative genomics. However, due to the intricate evolution of most gene families, which often involves multiple nested duplications, genomic rearrangements and horizontal gene transfers, orthology prediction remains as a highly challenging task (4,5), both analytically and computationally. Therefore multiple orthology resources have been developed that provide precomputed predictions, each based on a different methodology and organism range, and all having different strengths and weaknesses (6,7). The inference approaches fall into two main categories, namely graph-based (8–15) and tree-based (16–19) methods. Graph-based algorithms allow analysis of more species at once and produce groups of orthologous sequences with the common ancestor defined by the set of species considered at the taxonomic level. Tree-based approaches, by contrast, provide finer resolution (i.e. using tree topology to identify specific speciation and duplication events), but they require heavier computations and are more sensitive to methodological artifacts (20). We maintain a database of Orthologous Groups (OGs) and functional annotations called eggNOG (evolutionary genealogy of genes: Non-supervised Orthologous Groups) (21). eggNOG uses a graph-based unsupervised clustering algorithm extending the COG methodology (22) to produce genome wide orthology inferences, which are further adjusted to provide lineage specific resolution. The database currently covers 2031 eukaryotic and prokaryotic organisms, as well as precomputed mappings for 1655 additional prokaryotes (12). The present manuscript describes the most recent release of eggNOG (v4.5, 2015), featuring a number of improvements over its previous release. The most notable ones include (i) modifications to the clustering algorithm in order to make OGs hierarchically consistent across taxonomic levels, (ii) improved annotation of OGs, (iii) the availability of HMM-based tools for fast protein sequence assignment to OGs, (iv) the addition of viral OGs, (v) the availability of fine-grained orthology inferences derived from phylogenetic analysis, (vi) a completely re-designed web interface and (vii) programmatic access through a RESTful Application Programming Interface (API). eggNOG v4.5 is available at http://eggnog.embl.de. OVERVIEW OF THE COMPUTATIONAL PIPELINE Apart from the central graph-based clustering algorithm, the eggNOG production pipeline involves a number of quality controls as well as pre- and post-processing steps, which have evolved over the last 8 years since its first publication (21). Given the amount of change accumulated at present and previous versions, we describe here the current status of the complete pipeline (Figure 1), highlighting the most recent updates and additions since release 4.0 (12). [Figure: Schematic representation of the eggNOG pipeline: Boxes labelled in green indicate new data and/or methods added in this version. Blue labels represent updated methodology and/or data with respect to previous versions. Grey boxes indicate unchanged steps in version 4.5.] Data set preparation and pairwise sequence comparison The workflow starts by collecting genomes from public databases (19,23–26). Genomes and proteomes are downloaded, parsed and subjected to quality controls that prevent the inclusion of partial or draft genomes (Figure 1A). This step is coordinated with the STRING (27) and STITCH (28) databases so that the underlying set of protein sequences and names is shared among all three resources. The new viral proteins included in eggNOG v4.5 were retrieved by selecting all reference viral proteomes in Uniprot via XML download on 31 August 2015. These proteomes were filtered by a series of quality controls, which removed 50 proteomes. Eight additional proteomes were included following manual review. Viral proteins are often translated as a single polyprotein, which is cleaved to form functional proteins. Prior to inclusion in eggNOG, such polyproteins were cleaved in silico following the ‘chain’ annotations present in Uniprot entries, and only the smallest units were retained, so that the protein sequences are non-redundant. Pairwise sequence comparison Protein sequences from the selected organisms and viruses are extracted and used to compute an all-against-all pairwise similarity matrix (Figure 1B), a task that is currently carried out by the SIMAP project (29). The comparison uses Smith–Waterman alignments and compositional adjustment of the scores, as in BLAST, to prevent spurious hits between low-complexity sequence regions. Hits with bit-scores above 50 are stored and indexed in a relational database, which forms the input to the next stage of the algorithm. Definition of taxonomic levels Because the resolution of OGs depends on the taxonomic level, the eggNOG clustering pipeline is independently executed at different predefined taxonomic levels, each spanning a different clade in the overall tree of life. Levels are manually chosen to cover evolutionarily relevant groups as well as to maximally make use of well-studied model organisms (Figure 1C). This gives rise to the hierarchical structure of the data in eggNOG (Figure 2A), where, for example, a set of mammalian sequences with a common ortholog at the base of vertebrates could be part of a single mammal-specific OG (OG:0UIPS in Figure 2A), but constitute two separate supraprimate-specific groups (OG:1AVEH and OG:1AU76 in Figure 2A). In addition, eggNOG v4.5 uses 16 predefined taxonomic levels to classify the 352 viral proteomes (Figure 2B). [Figure: (A) Hierarchically consistent structure

Read the complete text