FunGene: the functional gene pipeline and repository.

Fish JA, Chai B, Wang Q, Sun Y, Brown CT, Tiedje JM, Cole JR. · 2013

Read FunGene: the functional gene pipeline and repository. by Fish JA, Chai B, Wang Q, Sun Y, Brown CT, Tiedje JM, Cole JR. (2013) 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-pmc3787254.txt.

Introduction

Jordan Fish and six co-authors described a resource called FunGene in Frontiers in Microbiology in 2013. Ribosomal RNA genes are the standard markers for studying microbial communities, but they have limits: they occur in multiple copies and change slowly, so distinct species can share identical sequences. Genes for key ecological jobs, such as carbon and nitrogen cycling, can reveal what a community does.

FunGene, the Functional Gene Pipeline and Repository, offers curated databases of common ecological functional genes and proteins. Its tools let researchers browse the collections, pick subsets, build family trees, test primer coverage, and download aligned sequences. A tool called FrameBot repairs frameshift errors in raw reads and finds the closest matching reference protein. The authors aim to give microbial ecology a direct view of function, not just identity.

Catalog record

CREATORFish JA, Chai B, Wang Q, Sun Y, Brown CT, Tiedje JM, Cole JR.
YEAR2013
COLLECTIONPAPERS
FORMATTXT
FILE SIZE72 KB
LICENSECC BY 4.0
RIGHTSCC BY
SOURCEEurope PMC

Source: https://europepmc.org/articles/PMC3787254

Browse the full library · About the Tower of Babel

Full text

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

FunGene: the functional gene pipeline and repository By Jordan A Fish, Benli Chai, Qiong Wang, Yanni Sun, C Titus Brown, James M Tiedje, James R Cole Published in: Frontiers in Microbiology DOI: 10.3389/fmicb.2013.00291 ABSTRACT Abstract Ribosomal RNA genes have become the standard molecular markers for microbial community analysis for good reasons, including universal occurrence in cellular organisms, availability of large databases, and ease of rRNA gene region amplification and analysis. As markers, however, rRNA genes have some significant limitations. The rRNA genes are often present in multiple copies, unlike most protein-coding genes. The slow rate of change in rRNA genes means that multiple species sometimes share identical 16S rRNA gene sequences, while many more species share identical sequences in the short 16S rRNA regions commonly analyzed. In addition, the genes involved in many important processes are not distributed in a phylogenetically coherent manner, potentially due to gene loss or horizontal gene transfer. While rRNA genes remain the most commonly used markers, key genes in ecologically important pathways, e.g., those involved in carbon and nitrogen cycling, can provide important insights into community composition and function not obtainable through rRNA analysis. However, working with ecofunctional gene data requires some tools beyond those required for rRNA analysis. To address this, our Functional Gene Pipeline and Repository (FunGene; http://fungene.cme.msu.edu/) offers databases of many common ecofunctional genes and proteins, as well as integrated tools that allow researchers to browse these collections and choose subsets for further analysis, build phylogenetic trees, test primers and probes for coverage, and download aligned sequences. Additional FunGene tools are specialized to process coding gene amplicon data. For example, FrameBot produces frameshift-corrected protein and DNA sequences from raw reads while finding the most closely related protein reference sequence. These tools can help provide better insight into microbial communities by directly studying key genes involved in important ecological processes. Keywords: microbial ecology, functional genes, amplification primers, phylogeny, biogeochemical cycles, amplicon analysis INTRODUCTION Microbes are a major component of the earth's biosphere. They impact the earth's ecosystems through a diverse set of activities and constitute a major force in shaping both abiotic and biotic environments. Examining the prevalence and diversity of the genes involved in these important ecological processes—ecofunctional genes—can help us define the relationships between microbial populations and their environments. Amplicon-based analysis of 16S rRNA gene fragments has shown great utility in providing an overall picture of taxonomic diversity and has been used in a multitude of environments (see, e.g., Gilbert and Meyer, 2012). However, much of the microbial gene content, including pathways for important environmental activities, have been subject to multiple horizontal gene transfers (see, e.g., Dagan et al., 2008). Directly querying for specific functional genes in environmental samples, through qPCR and amplicon surveys, can provide information on the prevalence and diversity of key functional genes regardless of discontinuity between organism and gene phylogeny. In contrast, those core protein-coding genes that are less subject to horizontal transfer and have gene phylogenies congruent with organism taxonomy can serve as important phylogenetic markers. These are often capable of resolving finer differences than the commonly used rRNA gene markers. Our concept of a microbial species is still evolving and a subject of debate, but the most commonly accepted ad hoc microbial species boundary has been 70% DNA-DNA re-association (Stackebrandt et al., 2002). Using this 70% re-association species definition, Stackebrandt and Ebers found that strains of the same species almost always share more than 99% full-length 16S rRNA gene identity, while different species can share identical 16S gene sequences (Stackebrandt and Ebers, 2006). The short 16S gene fragments commonly used in environmental amplicon analysis have even less resolving power. In contrast, the corresponding overall ortholog Average Amino Acid Identity (AAI; see list of abbreviations, Table 1) cutoff for a species is about 5% difference (Konstantinidis and Tiedje, 2005). This may vary for different marker genes, but in general protein-coding markers are capable of finer-grained resolution than rRNA genes (Santos and Ochman, 2004). [Table: List of Abbreviations.] Several toolkits for analyzing 16S data are currently available; examples include the RDP's Pyrosequencing Pipeline (Cole et al., 2009), QIIME (Caporaso et al., 2010), CloVr (Angiuoli et al., 2011), mothur (Schloss et al., 2009), and VAMPS (http://vamps.mbl.edu/). However, while parts of published 16S pipelines can be reused in functional gene pipelines, functional gene analysis presents several challenges not seen in 16S rRNA analysis. Indel sequencing errors cause frameshifts that lead to garbled protein translations. Few tools offer the functional gene reference sets required for analysis. FunFrame (Weisman et al., 2013) is an R-based analysis pipeline for functional gene data, built on analysis tools including HMMFrame (Zhang and Sun, 2011) for frameshift correction and gene translation. However, FunFrame comes customized for only one gene, Cytochrome D1. Sequences for different functional gene families can be analyzed using a common set of tools; however, each functional gene requires its own set of reference sequences and requires different approaches tuned to take into account factors such as relative gene diversity and amplicon size. Protein databases such as Pfam (Punta et al., 2012), TIGERFAM (Haft et al., 2003), and Uniprot (Uniprot Consortium, 2010) contain well-curated protein family sequence sets, however, they mainly concentrate on protein not gene sequences. In addition, the protein families are often defined more broadly than optimal for ecological analysis, for example covering paralogous genes active on different substrates. None of these general protein databases provide the tools necessary for functional gene amplicon analysis. The RDP's functional gene repository and analysis tools are available on two related websites. The FunGene Repository (FGR; http://fungene.cme.msu.edu/) is designed to help researchers explore publicly available sequences harvested from GenBank (Benson et al., 2004) and assigned to a specific collection of gene families. The FunGene Pipeline (FGP) houses tools for researchers to process and analyze their own functional gene sequencing data. FGR employs a reference/model-based, comparative analysis strategy to build the reference database and help to study functional and phylogenetic diversities of specific gene families. This strategy relies on the use of HMMER3 (http://hmmer.org/) and Hidden Markov Models (HMM), the same tool used by the well-known Pfam for studying protein homology. HMMs are built from protein seed alignments, i.e., typically a small number of vetted, full-length representative sequences from each gene family aligned by hand or with a Multiple Sequence Alignment tool [MSA; e.g., ClustalW (Larkin et al., 2007)]. Using these models, more inclusive sequence sets are created for the FGR by extracting, classifying, and aligning sequences obtained from GenBank. A model-based alignment method was chosen for several reasons. With a model-based alignment, multiple sets of alignments from a single model can be merged and compared without re-aligning the sequence database. Model-based approaches allow for statistical classification of query reads as members/not members of the modeled gene family. Aligning sequences with an HMM scales linearly in time with the number of sequences and is fast enough to align, within reasonable time, hundreds of thousands to millions of sequences. With HMM-based alignments, the conservation statistics of amino acid residues, essential to protein structure and function, can be preserved. This alignment is then back-translated to provide a high-quality nucleotide alignment, allowing more accurate comparative analysis at the nucleotide level based on sequence homology. Using FGP, libraries of sequence

Read the complete text