COPLA, a taxonomic classifier of plasmids

Abstract

Background: Plasmids are mobile genetic elements, key in the dissemination of antibiotic resistance, virulence determinants and other adaptive traits in bacteria. Obtaining a robust method for plasmid classification is necessary to better understand the genetics and epidemiology of many pathogens. Until now, plasmid classification systems focused on specific traits, which limited their precision and universality. The definition of plasmid taxonomic units (PTUs), based on average nucleotide identity metrics, allows the generation of a universal plasmid classification scheme, applicable to all bacterial taxa. Here we present COPLA, a software able to assign plasmids to known and novel PTUs, based on their genomic sequence.

Results: We implemented an automated pipeline able to assign a given plasmid DNA sequence to its cognate PTU, and assessed its performance using a sample of 1000 unclassified plasmids. Overall, 41% of the samples could be assigned to a previously defined PTU, a number that reached 63% in well-known taxa such as the Enterobacterales order. The remaining plasmids represent novel PTUs, indicating that a large fraction of plasmid backbones is still uncharacterized.

Conclusions: COPLA is a bioinformatic tool for universal, species-independent, plasmid classification. Offered both as an automatable pipeline and an open web service, COPLA will help bacterial geneticists and clinical microbiologists to quickly classify plasmids.

Keywords: Antibiotic resistance genes; Average nucleotide identity; Horizontal gene transfer; Plasmid; Plasmid epidemiology.

Fig. 1

Score distribution for 1000 plasmids sampled from RefSeq200, not present in the COPLA reference database (RefSeq84). The figure displays a semilogarithmic plot of the number of plasmids resulting in each given score

Fig. 2

Representative prediction outcomes. The query plasmid is represented by the node with the red inner circle. For all other nodes, the color of the inner circle represents the PTU assigned in the reference database (i.e. using only RefSeq84 plasmids). The outer ring colors represent the PTU assigned by COPLA. Yellow represents the PTU assigned to the query, green corresponds to nodes belonging to a different PTU, and grey represents not assigned PTUs. Case 1: the query represents a singleton. Case 2: the query belongs to a cluster with one or two members. A PTU cannot be assigned. Case 3: the query belongs to a cluster with three members. COPLA predicts a “new putative PTU”. Case 4: the query links together isolated plasmids to organize a 4-member cluster. COPLA predicts a “new putative PTU”. Case 5: the query clusters with the members of a known PTU. COPLA predicts that query belongs to that PTU. Case 6: the query links peripherally to a cluster corresponding to a known PTU. However, either the number of connections is not enough to fulfill the intercluster density rule, or the size of the query is not compatible to that of the PTU (see “Building the PTU reference catalog” in Implementation). COPLA output indicates that no PTU can be assigned to the query. Case 7: the query links peripherally to a cluster corresponding to a known PTU. The query organizes a subcluster of four members that does not fulfill the rules to integrate in the PTU. COPLA output predicts a “new putative PTU”. Case 8: As in case 7, the query organizes a subcluster that does not fulfill the rules to integrate in the PTU. Furthermore, it drags one member of the PTU to the new cluster. COPLA output predicts a “new putative PTU”. Case 9: the query significantly alters the structure of a known PTU. COPLA output predicts a “new putative PTU”. It also warns that “query is related to PTU-… plasmids”. See additional details and explanations in the main text (Discussion)