Rapid phylogenetic analysis of large samples of recombinant bacterial whole genome sequences using Gubbins.
Croucher NJ, Page AJ, Connor TR, Delaney AJ, Keane JA, Bentley SD, Parkhill J, Harris SR. · 2015
Read Rapid phylogenetic analysis of large samples of recombinant bacterial whole genome sequences using Gubbins. by Croucher NJ, Page AJ, Connor TR, Delaney AJ, Keane JA, Bentley SD, Parkhill J, Harris SR. (2015) 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-pmc4330336.txt.
Introduction
Nicholas J. Croucher and seven co-authors present Gubbins, a program that builds bacterial family trees while stripping out swapped DNA. Their paper appeared in Nucleic Acids Research in 2015. Whole genome data sets for bacteria keep growing, and recombination between strains piles dense clusters of changes onto some branches, which distorts a tree built naively.
Gubbins scans the alignment for stretches with unusually dense substitutions, the fingerprints of horizontal transfer, and builds the tree from the point mutations outside them. It repeats the pass until the tree settles. Simulations under realistic settings gave highly accurate trees. Outbreak investigators and evolution researchers use the output to trace how strains recently diverged.
Catalog record
| CREATOR | Croucher NJ, Page AJ, Connor TR, Delaney AJ, Keane JA, Bentley SD, Parkhill J, Harris SR. |
|---|---|
| YEAR | 2015 |
| COLLECTION | PAPERS |
| FORMAT | TXT |
| FILE SIZE | 68 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:
Rapid phylogenetic analysis of large samples of recombinant bacterial whole genome sequences using Gubbins By Nicholas J. Croucher, Andrew J. Page, Thomas R. Connor, Aidan J. Delaney, Jacqueline A. Keane, Stephen D. Bentley, Julian Parkhill, Simon R. Harris Published in: Nucleic Acids Research DOI: 10.1093/nar/gku1196 ABSTRACT The emergence of new sequencing technologies has facilitated the use of bacterial whole genome alignments for evolutionary studies and outbreak analyses. These datasets, of increasing size, often include examples of multiple different mechanisms of horizontal sequence transfer resulting in substantial alterations to prokaryotic chromosomes. The impact of these processes demands rapid and flexible approaches able to account for recombination when reconstructing isolates’ recent diversification. Gubbins is an iterative algorithm that uses spatial scanning statistics to identify loci containing elevated densities of base substitutions suggestive of horizontal sequence transfer while concurrently constructing a maximum likelihood phylogeny based on the putative point mutations outside these regions of high sequence diversity. Simulations demonstrate the algorithm generates highly accurate reconstructions under realistically parameterized models of bacterial evolution, and achieves convergence in only a few hours on alignments of hundreds of bacterial genome sequences. Gubbins is appropriate for reconstructing the recent evolutionary history of a variety of haploid genotype alignments, as it makes no assumptions about the underlying mechanism of recombination. The software is freely available for download at github.com/sanger-pathogens/Gubbins, implemented in Python and C and supported on Linux and Mac OS X. INTRODUCTION Many different patterns of genetic variation are evident within important microbial pathogens (1). The simplest cases are the genetically monomorphic, or ‘clonal’, species, such as Mycobacterium tuberculosis (2), which typically only accumulate variation through point mutation. Similar patterns of variation are found across much of the genome of species such as Salmonella enterica and Staphylococcus aureus, with horizontal exchange of sequence usually limited to the movement of mobile genetic elements (MGEs) (3,4). Nevertheless, rare but dramatic recombinations have been recorded outside of MGEs within these species, such as those between S. enterica serovars Typhi and Paratyphi A (5) and involved in the generation of the chimeric S. aureus sequence type (ST) 239 (6). However, such ‘homologous recombinations’ are far more common in naturally transformable species, such as Haemophilus influenzae, Neisseria meningitidis and Streptococcus pneumoniae (7). These species possess specific systems that actively import DNA from the environment and subsequently facilitate integration of the acquired fragments into their genome at a region of similar sequence. In such species, diversification often occurs through a complex combination of point mutation, homologous recombination and the movement of MGEs. The substantial lengths of sequence transferred via different recombination mechanisms can influence genome-wide measures of sequence similarity to a far greater extent than the vertically-inherited point mutations that are the signal of shared common ancestry (8). Hence identifying recombinations, and accounting for their potentially distinctive phylogenetic history, is crucial for tracing the evolution of many important human pathogens. Historically, two approaches have been taken for the detection of recombination events. The first attempts to find ‘exchanges’ within a sample; these methods search for evidence of unexpected similarity between divergent sequences, which are hypothesized to represent the movement of sequence from a donor into a recipient, or a common donor acting as a source of the sequence to both independently. Such exchanges result in homoplasies, which can cause difficulties in identifying the correct phylogenetic topology. One of the first attempts to use such an approach to detect recombination in protein sequences was an investigation of cytochrome evolution (9). Early examples using DNA sequences include tests for ‘incongruent phylogenetic partitions’ (10), distributions of discordant sites (11) and the ‘homoplasy test’ (12). More recent algorithms of this type, such as RDP3 (13), cBrother (14) and GARD (15), are able to identify the breakpoints in putative recombinant sequences based on the relative divergence between multiple representatives of a population in different regions of a sequence alignment. These methods are most effective when sequences are available for donor, recipient and recombinant genotypes. This is frequently not possible in bacteria owing to the substantial unsampled diversity of potential donors, which may be members of a different species or no longer extant in the population. Hence a second approach, which involves searching closely-related isolates for evidence of ‘imports’ of divergent sequence from a distantly related source, has proved valuable in the analysis of bacterial sequences. Necessarily, such methods are restricted to the study of subsets of bacterial ‘species’ (using the definition of a set of isolates that frequently exchange sequence (16)) as recombinations are identified as regions containing a high density of polymorphisms relative to the background level. One of the first such methods was the application of the maximum χ2 test to mosaic bacterial genes (17), which identified boundaries between regions of different single nucleotide polymorphism (SNP) density. Such acquisitions of diverse sequence have a dramatic impact on branch length (and hence clock rate estimation), and can also affect tree topology if convergence arises by similar sequences being acquired at the same genomic location in parallel on different branches. The study of recombination in closely-related bacteria led to the concept of a ‘clonal frame’ (18): the portion of a sequence that has not undergone recombination since a set of isolates diverged from their most recent common ancestor. This concept has proved valuable in tracing the evolutionary history of sets of similar bacterial isolates, such as observed in disease outbreaks. The most sophisticated software available to implement this approach is ClonalFrame (19), which uses a Bayesian Monte Carlo Markov chain (MCMC) approach to detect recombinations as regions with a significantly elevated density of polymorphisms while simultaneously constructing a tree based on the point mutations outside of the recombinations. This software has proved popular in the analysis of large collections of genotyping data (20) and smaller collections of whole genome sequences (21). However, the recent proliferation of second and third generation sequencing technologies has led to a rapid rise in the quantity of bacterial whole genome sequences available for analysis (22), resulting in a significant increase in the computation time taken to fit complex evolutionary models to such rich datasets. Hence, faster algorithms are necessary for the analysis of such samples; one example is BRATNextGen, which detects recombinations, but does not return a phylogenetic reconstruction of the sequences’ past history (23). Here, we present a rapid method for the detection of sequence imports through recombination in bacterial populations. It can be applied to large samples of closely-related haploid sequences in order to generate a prediction of putative recombination events and a phylogeny based on the presumed base substitutions in the ‘clonal frame’. It does not depend upon recombination donors being present in the sample, or require a set of recombination breakpoints to be predefined, a situation for which there is currently a dearth of alternatives capable of processing large genomic datasets (24). The basic statistical framework has already been applied to S. pneumoniae (25–30), Vibrio cholerae (31), Mycobacterium abscessus (32), S. enterica serovars Typhimurium (33,34) and Bovismorbificans (35), Escherichia coli (36) and Neisseria gonorrhoeae (37), as well as being adapted for an analysis of Chlamydia trachomatis (38). This work describes and evaluates a novel implementation in C, called Gubbins (Genealogies Unbiased By recomBinations In Nucleotide Sequences), which represents the