RDP5: a computer program for analyzing recombination in, and removing signals of recombination from, nucleotide sequence datasets.

Martin DP, Varsani A, Roumagnac P, Botha G, Maslamoney S, Schwab T, Kelz Z, Kumar V, Murrell B. · 2021

Read RDP5: a computer program for analyzing recombination in, and removing signals of recombination from, nucleotide sequence datasets. by Martin DP, Varsani A, Roumagnac P, Botha G, Maslamoney S, Schwab T, Kelz Z, Kumar V, Murrell B. (2021) 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-pmc8062008.txt.

Introduction

Martin and co-authors present RDP5, a program for finding recombination in nucleotide sequence datasets. Recombination scrambles the signal that evolution studies rely on. For 20 years, the RDP project has built tools to detect it. Earlier versions leaned on users to check each event by hand. RDP5 automates that checking well enough to run inside pipelines.

New statistical tests weed out signals that mimic recombination but arise from other processes. Speed gains offset the extra work of those tests. The authors report runs up to five times faster than RDP4. The program handles alignments of up to 5,000 sequences and 50 million sites. It can also strip recombination out of a dataset entirely. Cleaned alignments come out in many formats, ready for downstream work.

Catalog record

CREATORMartin DP, Varsani A, Roumagnac P, Botha G, Maslamoney S, Schwab T, Kelz Z, Kumar V, Murrell B.
YEAR2021
COLLECTIONPAPERS
FORMATTXT
FILE SIZE14 KB
LICENSECC BY 4.0
RIGHTSCC BY
SOURCEEurope PMC

Source: https://europepmc.org/articles/PMC8062008

Browse the full library · About the Tower of Babel

Full text

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

RDP5: a computer program for analyzing recombination in, and removing signals of recombination from, nucleotide sequence datasets By Darren P Martin, Arvind Varsani, Philippe Roumagnac, Gerrit Botha, Suresh Maslamoney, Tiana Schwab, Zena Kelz, Venkatesh Kumar, Ben Murrell Published in: Virus Evolution DOI: 10.1093/ve/veaa087 ABSTRACT Abstract For the past 20 years, the recombination detection program (RDP) project has focused on the development of a fast, flexible, and easy to use Windows-based recombination analysis tool. Whereas previous versions of this tool have relied on considerable user-mediated verification of detected recombination events, the latest iteration, RDP5, is automated enough that it can be integrated within analysis pipelines and run without any user input. The main innovation enabling this degree of automation is the implementation of statistical tests to identify recombination signals that could be attributable to evolutionary processes other than recombination. The additional analysis time required for these tests has been offset by algorithmic improvements throughout the program such that, relative to RDP4, RDP5 will still run up to five times faster and be capable of analyzing alignments containing twice as many sequences (up to 5000) that are five times longer (up to 50 million sites). For users wanting to remove signals of recombination from their datasets before using them for downstream phylogenetics-based molecular evolution analyses, RDP5 can disassemble detected recombinant sequences into their constituent parts and output a variety of different recombination-free datasets in an array of different alignment formats. For users that are interested in exploring the recombination history of their datasets, all the manual verification, data management and data visualization components of RDP5 have been extensively updated to minimize the amount of time needed by users to individually verify and refine the program’s interpretation of each of the individual recombination events that it detects. 1. INTRODUCTION Recombination and genome component reassortment are processes that strongly impact the evolution of many virus species. The patterns of nucleotide sequence variation created in virus genomic sequence datasets by these processes (which are hereafter collectively referred to as recombination) can seriously undermine the accuracy of phylogenetics-based molecular evolution analyses (Schierup and Hein 2000a,b; Scheffler, Martin, and Seoighe 2006; Arenas and Posada 2010). It is therefore frequently desirable to test nucleotide sequence datasets for evidence of recombination and, when such evidence is found, take steps to minimize the impacts of detected recombination events on downstream analyses of these datasets. Of the many available computer programs for analyzing recombination (see http://bioinf.man.ac.uk/robertson/recombination/programs.shtml; Martin et al. 2011), recombination detection program (RDP) is presently one of the most commonly used. Successive versions of RDP have applied an expanding array of recombination event detection, recombination breakpoint demarcation, and recombinant sequence identification methods, all applied in unison, to yield detailed descriptions of how recombination may have impacted the evolution of any given set of aligned nucleotide sequences (Martin et al. 2015). The accuracy of these descriptions, however, frequently depended on the amount of effort users were willing to put into exploring the many similarly plausible ways in which the detected patterns of recombination may have arisen. A guiding principle during the development of RDP5, the latest version of the RDP series, has therefore been a minimization of the amount of time that users need to invest in detecting and removing signals of recombination from nucleotide sequence datasets. 2. GENERATION OF RECOMBINATION-FREE DATASETS RDP5 can take as input an aligned nucleotide sequence dataset, automatically identify and characterize individual recombination events that are evident within that dataset, and output modifications of the input alignment within which all signals of detectable recombination have been removed. The different types of modified recombination-free multiple sequence datasets that RDP5 can output include alignments where: (1) recombinant sequences have been removed; (2) fragments of sequence derived through recombination have been removed; (3), recombinant sequences are split up into their constituent parts; and (4) the input alignment is divided into multiple different gene/genome sub-region alignments based on the locations of detected recombination breakpoints (Supplementary Fig. S1). 3. QUERY VS REFERENCE SCANS FOR RECOMBINATION In addition to the fully exploratory recombination analysis modes found in previous versions of the program (Martin et al. 2015), RDP5 also includes a new highly automated ‘query vs reference’ analysis mode such as that found in the programs REGA (de Oliveira et al. 2005) and jpHMM (Schultz et al. 2006). Unlike with the default fully exploratory recombination analysis mode where every sequence is tested for evidence of recombination, the query vs reference mode will test a user-defined set of query sequences for evidence that they originated through recombination between a user-defined set of reference sequences. Such an analysis mode is well suited to analyzing patterns of recombination between two or more groups of viruses that have only recently had the opportunity to start recombining with one another: such as in an individual patient that has been infected with two distinct variants of a virus (Sheward et al. 2018), or within a geographical region where multiple distinct genetic variants of a virus have recently started co-circulating. To minimize the amount of effort needed to define reference sequences and/or groups of reference sequences, RDP5 will automatically identify sequences as queries and references based on simple sequence naming rules (see the manual provided with RDP5 for these rules). These naming rules can also be used to group reference sequences into different reference types. 4. AUTOMATED SEQUENCE ANNOTATION Given an accessible internet connection, RDP5 will automatically annotate the genomic features of input virus sequence datasets using the curated NCBI virus reference sequence database (https://www.ncbi.nlm.nih.gov/genome/viruses/). Annotation is useful in the context of generating recombination-free datasets because it enables RDP5 to output gene sequence alignments that are suitable for downstream codon-focused analyses of natural selection. The gene sequence alignments that RDP5 produces can account for variation in the positions of gene start and stop codons, intron splicing (wherever intron donor and acceptor sites are annotated in the NCBI reference sequence records), portions of genes that are expressed in two or more different open reading frames (these regions can be excluded) and real and/or artifactual frame-shift mutations (partial codons can be removed). Sequence annotation is also useful for testing the selective processes that impact patterns of recombination detected within sequence datasets. Given a set of annotated genome sequences recombination breakpoint distribution tests in RDP5 will now automatically detect whether observed recombination breakpoint distributions vary between: (1) coding and non-coding regions; (2) different genes; and (3) the edges and internal regions of genes. 5. DETECTION OF POTENTIAL FALSE-POSITIVE RECOMBINATION SIGNALS A crucial facilitator of the highly automated recombination analyses that RDP5 can perform is the inclusion of tests that are specifically designed to detect and flag potential false-positive signals of recombination. Besides automatically testing whether each of the recombination signals that RDP5 detects might be attributable to sequence misalignment (which is a major contributor to false positive signals of recombination), the program will also detect and flag as suspicious any detectable recombination signals that may have arisen through evolutionary processes other than recombination. Specifically, RDP5 uses the PHI test (Bruen, Philippe, and Bryan 2006) and 4-gamete test (McVean, Awadalla, and Fearnhead 2002) adapted versions of the homoplasy test (Maynard Smith and Smith 1998) to flag apparent recombination signals that are potentially attributable to

Read the complete text