Please use this identifier to cite or link to this item:
logo share SHARE BASE
Visualizar otros formatos: MARC | Dublin Core | RDF | ORE | MODS | METS | DIDL | DATACITE

Accnet: New tool for accessory genome analysis using bipartite networks

AuthorsLanza, Val F. CSIC; Rodriguez, Irene; Tedim, Ana P.; Toro, María de; Cruz, Fernando de la CSIC ORCID; Cantón, Rafael; Baquero, Fernando; Coque, Teresa M.
Issue Date2016
PublisherEuropean Society of Clinical Microbiology and Infectious Diseases
CitationXI IMMEM (2016)
Abstract[Background]: Lateral gene transfer plays an essential role in the bacterial evolution The genomes of most opportunistic pathogens differ in their accessory genome which include determinants associated with virulence, colonization or antibiotic resistance, among other adaptive traits. It exists a wide range of bioinformatics tools for typing, core genome analysis or phylogenetics but only a few tools for analysing the accessory genomes. We present AccNET (Accesory Constelation Network)), a new tool, non-computational expensive, based on bipartite networks that is able to analyse simultaneously tens of genomes.[Material & Methods]: AccNET uses the kClust software to customly detect highly-related homologous genes, by modifying the identity threshold. The core genome, defined as the subset of proteins present in all genomes, is removed from the analysis. Then, AccNET builds a bipartite network of the accessory genome with two kind of nodes: genomes and proteins. The edges that link the nodes represent the presence of a protein in a given genome. The edge-weight is determined as a phylogenetic distance between the proteins included in the cluster. To calculate such distance, we use a pipeline: First, we align all the proteins belonged to the cluster using MUSCLE. Then, we optimize the multiple alignment using Trimal and finally we calculate the distance matrix using ProtDist, (PHYLIP package). For each protein the phylogenetic distance for edge-weight is determined by the mean of the distance with all of the remaining proteins. Final output network is compatible with most of the network viewer software such as Cytoscape or Gephi. We used ForceAtlas2 layout from Gephi to arrange networks. To compare the structure of the accessory distribution with the phylogenetic relationships, we build the core-genome phylogenetic tree using the previous defined core-genome. First, we aligned each one of the core genes using MUSCLE. Then, we concatenated all the multiple alignments and extract the SNPs using HarvestTools. Finally, we built a Maximum Likelihood phylogenetic tree using FastTree2 with the resulting SNPs. To demonstrate AccNET capabilities, we have used two datasets that comprises genomes of the polyclonal group ST131 of E. coli (n=37) and the phylogenomic group BAPS 2.1 of E. faecium (n=25). Both include specific clonal variants worldwide disseminated. [Results]: The resulting network from E. coli dataset comprises 6973 nodes and 41585 edges while that of E. faecium has 3326 nodes and 20041 edges. A high correlation was noted between the clusters in the network of accessory genomes and the branches in the phylogenetic tree of the core genomes. However, the networks disclose specific interactions between genome clusters that are remote in the phylogenetic tree. Besides, both network shown specific protein clusters that belong to genome clusters associated with epidemic multidrug resistant variants such as E. coli ST131-fimH30 or E. faecium ST117. Fig1 show a snapshot of the network arrangements obtained by using AccNET and the core-phylogenetic tree applied to the E. coli dataset. [Conclusion]: We have developed an specific software tool for the analysis of accessory genome. This software is able to analysis tens of genomes with high precision. User friendly environment as Cytoscape allow us comprehensively exploring the accessory genome, and identifying specific genes/gene clusters of interest in evolutionary biology and epidemiological surveillance.
DescriptionResumen del póster presentado al 11th International Meeting on Microbial Epidemiological Markers, celebrado en Estoril (Portugal) del 9 al 12 de marzo de 2016.
Appears in Collections:(IBBTEC) Comunicaciones congresos

Files in This Item:
File Description SizeFormat
accesoRestringido.pdf15,38 kBAdobe PDFThumbnail
Show full item record
Review this work

Google ScholarTM


WARNING: Items in Digital.CSIC are protected by copyright, with all rights reserved, unless otherwise indicated.