BPGA- an ultra-fast pan-genome analysis pipeline.

Chaudhari NM, Gupta VK, Dutta C. · 2016

Read BPGA- an ultra-fast pan-genome analysis pipeline. by Chaudhari NM, Gupta VK, Dutta C. (2016) 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-pmc4829868.txt.

Introduction

Narendrakumar M. Chaudhari, Vinod Kumar Gupta, and Chitra Dutta present a pan-genome pipeline in Scientific Reports in 2016. A species' pan genome has three parts. Core genes are shared by all strains. Accessory genes are carried by some. Unique genes belong to single strains. That map exposes diversity and evolution. Older software was slow, fussy to install, or thin in features.

Their tool, BPGA, runs the standard analysis at high speed. It adds seven modules. These build family trees from the core or the whole set. They check for genes special to chosen strains. They run subset analysis and base-composition checks. They map each gene class to its functions. The tool asks little of the computer. It offers a plain command line and clean graphics. A test on 28 strains of a strep species shows it off.

Catalog record

CREATORChaudhari NM, Gupta VK, Dutta C.
YEAR2016
COLLECTIONPAPERS
FORMATTXT
FILE SIZE38 KB
LICENSECC BY 4.0
RIGHTSCC BY
SOURCEEurope PMC

Source: https://europepmc.org/articles/PMC4829868

Browse the full library · About the Tower of Babel

Full text

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

BPGA- an ultra-fast pan-genome analysis pipeline By Narendrakumar M Chaudhari, Vinod Kumar Gupta, Chitra Dutta Published in: Scientific Reports DOI: 10.1038/srep24373 ABSTRACT Abstract Recent advances in ultra-high-throughput sequencing technology and metagenomics have led to a paradigm shift in microbial genomics from few genome comparisons to large-scale pan-genome studies at different scales of phylogenetic resolution. Pan-genome studies provide a framework for estimating the genomic diversity of the dataset, determining core (conserved), accessory (dispensable) and unique (strain-specific) gene pool of a species, tracing horizontal gene-flux across strains and providing insight into species evolution. The existing pan genome software tools suffer from various limitations like limited datasets, difficult installation/requirements, inadequate functional features etc. Here we present an ultra-fast computational pipeline BPGA (Bacterial Pan Genome Analysis tool) with seven functional modules. In addition to the routine pan genome analyses, BPGA introduces a number of novel features for downstream analyses like core/pan/MLST (Multi Locus Sequence Typing) phylogeny, exclusive presence/absence of genes in specific strains, subset analysis, atypical G + C content analysis and KEGG & COG mapping of core, accessory and unique genes. Other notable features include minimum running prerequisites, freedom to select the gene clustering method, ultra-fast execution, user friendly command line interface and high-quality graphics outputs. The performance of BPGA has been evaluated using a dataset of complete genome sequences of 28 Streptococcus pyogenes strains. With the advent of high-throughput low-cost sequencing technologies and metagenomic techniques, the field of microbial genomics has experienced a paradigm shift from single or few genome analyses to large-scale comparisons of hundreds to thousands of genomes. One of the important outcomes of such multi-genome studies is the concept of the pan-genome - a term coined by Tettelin et al. in 2005 to refer to the complete inventory of genes in a particular species [in Greek, pan (παν) means ‘whole’]1. A study of seven Streptococcus agalactiae genomes by Tettelin et al. demonstrated that strains of a bacterial species might differ substantially in their gene content and total gene pool of a species might be orders of magnitudes larger than the gene content of any single strain1. It is, therefore, rational to describe a bacterial species by its pan-genome, which includes a core genome containing genes shared by all strains, a dispensable genome containing accessory genes that exist in two or more strains and genes unique to single strains (also known singletons). The core genes are responsible for the basic aspects of the biology of the species and its major phenotypic traits; while the accessory genes and singletons usually pertain to supplementary biochemical pathways and functions that may confer selective advantages such as ecological adaptation, virulence mechanisms, antibiotic resistance, or colonization of a new host. The pan genome analyses represent a new approach to species definition and provide a framework for estimating and/or modeling the genetic diversity of the study group. During last ten years, pan genomic studies have been conducted on nearly fifty bacterial species, which include model organisms like Escherichia coli and members of normal human flora like Lactobacillus paracasei as well as a number of pathogens like Haemophilus influenza, Coxiella burnetii, Streptococcus pneumonia, Yersinia pestis etc.2,3,4,5,6,7. Traditionally defined at the species level, the pan genomic approach has later been implemented also at higher levels of phylogenetic resolution ranging from genus to phylum and beyond. Genus level studies were carried out on Streptococcus, Salmonella, Prochlorococcus etc., and the pan genome study of the Chlamydiae is an example of phylum level studies. Recently, Lapierre & Gogarten has extended the concept of the pan genome to the entire bacterial domain8,9,10,11,12. The concept has also been implemented in viral, plant and fungal genome studies13,14,15,16,17. Pan genome analyses have provided valuable insight into genome dynamics, population structure, species evolution, niche specialization, pathogenesis, drug resistance and many other features of the microbial world1,18,19,20,21,22. It has also been exploited for development of vaccines against bacteria23. A number of free tools and web servers are available for pan genome analysis, but each of them suffers from one or the other limitations, leaving rooms for further improvement. For instance, PanOCT, being a web-based database, is applicable only to a limited number of species. Panseq, PanGP and Roary provide few functional features and hence, lack in downstream analyses. PGAP and ITEP include a number of functional modules, but are much slower in speed24,25,26,27,28,29. There has been, therefore, a pressing need for development a new computational pipeline, which will not only offer fast and efficient formalisms for construction of the pan genome through clustering of orthologous gene families and but also enable various downstream analyses such as mapping of the core, accessory & unique genes to various COG categories and/or KEGG pathways, phylogenetic analysis, in silico multi locus sequence typing (MLST) and other relevant analyses. The pipeline should also provide options for selecting from different tools available for orthologous clustering and formation of binary gene presence/absence matrix. An option for applying the tools to a subset of the total dataset may facilitate identification of exclusive genetic features that can discriminate between different serological, ecological or pathogenic groups. In this context; we have developed an ultra-fast computational pipeline BPGA (Bacterial Pan Genome Analysis Tool) with seven functional modules for comprehensive pan genome studies and downstream analyses. In addition to all types of analyses offered by currently available tools, this pipeline contains certain novel features like Exclusive Gene Family Analysis, KEGG Pathway Analysis, GC Content Analysis, Subset Analysis etc. Other notable features of BPGA includes minimum running prerequisites, ease of handling, user friendly command line interface, freedom for user to select method for clustering, high quality image output and efficiency in terms of time cost. METHODS BPGA is written in perl programing language but complied in executable files for both Windows and Linux so that no module installation is required. The perl code is freely available to make BPGA system independent. Dependencies like MUSCLE and rsvg-convert are also provided inside the installer, required for sequence alignments and tree generation30. Another prerequisite gnuplot 4.6.6 which is required for plotting results into graphs; should be manually installed on the system. BPGA performs pre-processing step to prepare sequence data for clustering. BPGA then runs USEARCH for fastest clustering (using 50% sequence identity cut-off by default; user may change this cut-off value). The clustered output is processed to generate tab delimited gene presence absence binary matrix (pan-matrix) which is then used for pan-genome profile calculations with iterations (default 20 or user defined) as well as pan genome based phylogeny. MUSCLE is used to align concatenated core genes to generate phylogeny tree based on core genome and MLST based on user selected housekeeping genes. For assigning COG and KEGG IDs, best hits with respective reference databases obtained from ublast function of USEARCH are used. Gnuplot is used for plotting all the results as high quality pdf images. Details about each step are provided in Table 1. [Table: Description of BPGA Pipeline.] BPGA Algorithm BPGA performs pan genome analysis in stepwise manner using its core modules. The steps include: Input preparation for clustering Three types of input files can be processed by BPGA: - GenBank file: BPGA processes GenBank (.gbk) files for orthologous cluster analysis and generates an input file that may be used as an

Read the complete text