Phamerator: a bioinformatic tool for comparative bacteriophage genomics.
Cresawn SG, Bogel M, Day N, Jacobs-Sera D, Hendrix RW, Hatfull GF. · 2011
Read Phamerator: a bioinformatic tool for comparative bacteriophage genomics. by Cresawn SG, Bogel M, Day N, Jacobs-Sera D, Hendrix RW, Hatfull GF. (2011) 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-pmc3233612.txt.
Introduction
Steven Cresawn and colleagues introduced Phamerator, software for comparing bacteriophage genomes, in this 2011 paper in BMC Bioinformatics. Phages, the viruses that infect bacteria, have mosaic genomes stitched from exchanged pieces and crowded with small genes of unknown function. That patchwork makes annotation and comparison unusually hard.
Phamerator sorts protein-coding genes into families of related sequences, called phamilies, using pairwise comparisons, and stores the relationships in a database. From it, researchers can draw genome maps of several phages at once that display both DNA and protein relationships plus conserved domains. Circular diagrams of each family reveal how individual genes have moved through phage populations by horizontal exchange, each with its own history. The tool gives phage biologists a shared picture of genomes that otherwise resist comparison.
Catalog record
| CREATOR | Cresawn SG, Bogel M, Day N, Jacobs-Sera D, Hendrix RW, Hatfull GF. |
|---|---|
| YEAR | 2011 |
| COLLECTION | PAPERS |
| FORMAT | TXT |
| FILE SIZE | 44 KB |
| LICENSE | CC BY 4.0 |
| RIGHTS | CC BY |
| SOURCE | Europe PMC |
Browse the full library · About the Tower of Babel
Full text
The opening of the work, from the library's hosted text:
Phamerator: a bioinformatic tool for comparative bacteriophage genomics By Steven G Cresawn, Matt Bogel, Nathan Day, Deborah Jacobs-Sera, Roger W Hendrix, Graham F Hatfull Published in: BMC Bioinformatics DOI: 10.1186/1471-2105-12-395 ABSTRACT Abstract Background Bacteriophage genomes have mosaic architectures and are replete with small open reading frames of unknown function, presenting challenges in their annotation, comparative analysis, and representation. Results We describe here a bioinformatic tool, Phamerator, that assorts protein-coding genes into phamilies of related sequences using pairwise comparisons to generate a database of gene relationships. This database is used to generate genome maps of multiple phages that incorporate nucleotide and amino acid sequence relationships, as well as genes containing conserved domains. Phamerator also generates phamily circle representations of gene phamilies, facilitating analysis of the different evolutionary histories of individual genes that migrate through phage populations by horizontal genetic exchange. Conclusions Phamerator represents a useful tool for comparative genomic analysis and comparative representations of bacteriophage genomes. BACKGROUND Bacteriophages represent a numerically vast, highly dynamic, evolutionarily ancient, and genetically highly diverse population [1-3]. Phage genomes are typically small compared to those of their bacterial hosts (ranging from a few to several hundred kilobases) and no longer present significant technical challenges to sequence determination [1]. As genomic sequencing approaches get simpler and cheaper, the availability of individual phage isolates for characterization becomes limiting, a need that can be effectively addressed through integrated research-education programs involving undergraduate and high school student investigators [4,5]. In spite of their relatively small size, phages present significant challenges to accurate genome annotation including gene identification. Two principal issues arise. First, phage genes tend on average to be small (~600 bp), approximately two-thirds the average size of bacterial genes [1,6]. Many of the genes required for virion structure and assembly are relatively large (tape measure genes can be over 6 kbp long), but those in the non-structural genomic segments are small, often shorter than 100 codons. Secondly, phage genomes are replete with genes of unknown function for which no homologues have been described [7-10]. Mosaic architectures are hallmarks of phage genomes, and individual phages can be considered as particular combinations of interchangeable modules, each of which can be present in two or more different genomic contexts [10,11]. In some cases, where the recombination events giving rise to these mosaic structures occurred relatively recently in evolutionary time, mosaicism is apparent through nucleotide sequence comparisons [12-14]. When the events occurred in more remote evolutionary times the evidence of common ancestry is usually no longer apparent at the nucleotide level, but often can be revealed from comparison of the predicted amino acid sequences [15-17]. Such comparisons reveal that individual phage genomes are typically constructed from multiple modules - often corresponding to single genes - each of which has a distinctly different phylogeny [10]. As such, accurate compilations of whole genome phylogenies that reflect the evolutionary history of the entire genome are not possible, and reticulate-based representations are needed to capture this evolutionary complexity [16,18]. The mechanisms giving rise to genome mosaicism are unclear but must accommodate the striking observation that module boundaries correspond closely with gene boundaries, and in some cases, domain boundaries [11,19]. One model invokes homologous recombination events targeted to short conserved boundary sequences between genes, and there is evidence for this in some phage genomes [20,21]. However, there are numerous examples where no conserved boundary sequences are evident, raising the possibility that mosaicism results largely from illegitimate recombination events between randomly chosen partners sharing little or no sequence identity [10,11]. In this second model, correspondence between module and gene boundaries results from the selection for gene function, not from targeting of the recombination events [22]. Comparison of genomes from phages that infect taxonomically diverse hosts typically provides little information into their evolution because of only very limited similarity at either the nucleotide or amino acid sequence level [4,11]. Phages of a common host, however, have the advantage that they are more likely to have been in recent genetic communication with each other and to have exchanged modules in recent evolutionary times [6]. Large sets of phage genomes are now available for several hosts including Burkholderia [9], Bacillus [23], Enterobacteriaceae [24], Mycobacteria [6], Prochlorococcus and Synechococcus [25], Pseudomonas [7], and Staphylococcus [8], although even these can span enormous genetic diversity [1,6]. A large number of phages that infect Mycobacterium smegmatis mc2155 have been isolated and a comparative analysis of 80 has been described [4,6,10,14]. Although these are genetically diverse, the diversity is heterogeneous, and phages can be grouped into 'clusters' according to their overall nucleotide sequence relationships [6]. Of the 80 published completely sequenced mycobacteriophage genomes, 75 can be grouped into ten major clusters, seven of which can be further subdivided into subclusters according to the extent of the nucleotide similarities [14]. Five of the genomes have no close relatives and are referred to as 'singletons' [6]. Because the currently sequenced mycobacteriophage genomes under-represent the mycobacteriophage population-at-large, these cluster designations will undergo modifications as new genomes are sequenced [14]. There are, however, numerous examples of genes that are shared between phages of different clusters and whose common ancestry is only apparent from amino acid sequence similarity [4,6,10]. We have proposed previously [4] that mycobacteriophage genes related to each other can be grouped into phamilies (phams) and that mosaic relationships can be analyzed and represented using pham-annotated genomes maps and phamily circles that show the patterns of which phages contain members of particular phams. Although manual or semi-automated approaches are applicable when only small numbers of genomes are analyzed [4], this becomes an impossible task as the number of genomes expands. We describe here a software program 'Phamerator' that provides bioinformatic tools for both analyzing and representing phage genome mosaicism. The core functionality of Phamerator performs pair wise similarity searches between predicted protein products of a set of phage genomes, and assorts them into phamilies (phams) of related sequences. Genome maps can be displayed that illustrate the relationships between phages at both the nucleotide and amino acid sequence level. Moreover, the evolutionary histories of specific genes can be displayed by phamily circles in which all gene members of particular phams are represented, and for which multiple phams can be compared. We illustrate the utilities of Phamerator using a set of 111 completely sequenced mycobacteriophage genomes, but the program is applicable to any set of phage genomes for which comparative analysis is desired. METHODS Phamerator database architecture Phamerator is written entirely in the Python computer programming language and makes use of a number of modules, including the Biopython framework for computational biology [26]. Biopython provides a programmatic interface for sequence manipulation, the construction and parsing of files in relevant formats, and access to external command line applications such as those used for sequence alignment. For the latter, Phamerator uses Biopython to interact with local instances of BLASTP and CLUSTALW. Due to the significant computational time required for performing large numbers of sequence alignments, Phamerator employs a distributed processing model that is implemented using Python Remote Objects (Pyro). Phamerator was developed on Ubuntu Linux but should be able to run on any modern UNIX-based operating system. Phamerator uses the MySQL database software with a simple, custom database schema that incorporates and extends the