BACKGROUND: Metagenomics experiments often make inferences about microbial communities by sequencing 16S and 18S rRNA, and taxonomic assignment is a fundamental step in such studies. This paper addresses the weaknesses in two types of metrics commonly used by previous studies for measuring the performance of existing taxonomic assignment methods: Sequence count based metrics and Binary error measurement. These metrics made performance evaluation results biased, less informative and mutually incomparable. RESULTS: We investigated weaknesses in two types of metrics and proposed new performance metrics including Average Taxonomy Distance (ATD) and ATD_by_Taxa, together with the visualized ATD plot. CONCLUSIONS: By comparing the evaluation results from four popular taxonomic assignment methods across three test data sets, we found the new metrics more robust, informative and comparable.

BACKGROUND: Estimating the parameters that describe the ecology of viruses,particularly those that are novel, can be made possible using metagenomic approaches. However, the best-performing existing methods require databases to first estimate an average genome length of a viral community before being able to estimate other parameters, such as viral richness. Although this approach has been widely used, it can adversely skew results since the majority of viruses are yet to be catalogued in databases. RESULTS: In this paper, we present ENVirT, a method for estimating the richness of novel viral mixtures, and for the first time we also show that it is possible to simultaneously estimate the average genome length without a priori information. This is shown to be a significant improvement over database-dependent methods, since we can now robustly analyze samples that may include novel viral types under-represented in current databases. We demonstrate that the viral richness estimates produced by ENVirT are several orders of magnitude higher in accuracy than the estimates produced by existing methods named PHACCS and CatchAll when benchmarked against simulated data. We repeated the analysis of 20 metavirome samples using ENVirT, which produced results in close agreement with complementary in virto analyses. CONCLUSIONS: These insights were previously not captured by existing computational methods. As such, ENVirT is shown to be an essential tool for enhancing our understanding of novel viral populations.

BACKGROUND: In metagenomics, the separation of nucleotide sequences belonging to an individual or closely matched populations is termed binning. Binning helps the evaluation of underlying microbial population structure as well as the recovery of individual genomes from a sample of uncultivable microbial organisms. Both supervised and unsupervised learning methods have been employed in binning; however, characterizing a metagenomic sample containing multiple strains remains a significant challenge. In this study, we designed and implemented a new workflow, Coverage and composition based binning of Metagenomes (CoMet), for binning contigs in a single metagenomic sample. CoMet utilizes coverage values and the compositional features of metagenomic contigs. The binning strategy in CoMet includes the initial grouping of contigs in guanine-cytosine (GC) content-coverage space and refinement of bins in tetranucleotide frequencies space in a purely unsupervised manner. With CoMet, the clustering algorithm DBSCAN is employed for binning contigs. The performances of CoMet were compared against four existing approaches for binning a single metagenomic sample, including MaxBin, Metawatt, MyCC (default) and MyCC (coverage) using multiple datasets including a sample comprised of multiple strains. RESULTS: Binning methods based on both compositional features and coverages of contigs had higher performances than the method which is based only on compositional features of contigs. CoMet yielded higher or comparable precision in comparison to the existing binning methods on benchmark datasets of varying complexities. MyCC (coverage) had the highest ranking score in F1-score. However, the performances of CoMet were higher than MyCC (coverage) on the dataset containing multiple strains. Furthermore, CoMet recovered contigs of more species and was 18 - 39% higher in precision than the compared existing methods in discriminating species from the sample of multiple strains. CoMet resulted in higher precision than MyCC (default) and MyCC (coverage) on a real metagenome. CONCLUSIONS: The approach proposed with CoMet for binning contigs, improves the precision of binning while characterizing more species in a single metagenomic sample and in a sample containing multiple strains. The F1-scores obtained from different binning strategies vary with different datasets; however, CoMet yields the highest F1-score with a sample comprised of multiple strains.

MOTIVATION: The combined effect of a high replication rate and the low fidelity of the viral polymerase in most RNA viruses and some DNA viruses results in the formation of a viral quasispecies. Uncovering information about quasispecies populations significantly benefits the study of disease progression, antiviral drug design, vaccine design and viral pathogenesis. We present a new analysis pipeline called ViQuaS for viral quasispecies spectrum reconstruction using short next-generation sequencing reads. ViQuaS is based on a novel reference-assisted de novo assembly algorithm for constructing local haplotypes. A significantly extended version of an existing global strain reconstruction algorithm is also used. RESULTS: Benchmarking results showed that ViQuaS outperformed three other previously published methods named ShoRAH, QuRe and PredictHaplo, with improvements of at least 3.1-53.9% in recall, 0-12.1% in precision and 0-38.2% in F-score in terms of strain sequence assembly and improvements of at least 0.006-0.143 in KL-divergence and 0.001-0.035 in root mean-squared error in terms of strain frequency estimation, over the next-best algorithm under various simulation settings. We also applied ViQuaS on a real read set derived from an in vitro human immunodeficiency virus (HIV)-1 population, two independent datasets of foot-and-mouth-disease virus derived from the same biological sample and a real HIV-1 dataset and demonstrated better results than other methods available.

Estimating the number of different species (richness) in a mixed microbial population has been a main focus in metagenomic research. Existing methods of species richness estimation ride on the assumption that the reads in each assembled contig correspond to only one of the microbial genomes in the population. This assumption and the underlying probabilistic formulations of existing methods are not useful for quasispecies populations where the strains are highly genetically related. The lack of knowledge on the number of different strains in a quasispecies population is observed to hinder the precision of existing Viral Quasispecies Spectrum Reconstruction (QSR) methods due to the uncontrolled reconstruction of a large number of in silico false positives. In this work, we formulated a novel probabilistic method for strain richness estimation specifically targeting viral quasispecies. By using this approach we improved our recently proposed spectrum reconstruction pipeline ViQuaS to achieve higher levels of precision in reconstructed quasispecies spectra without compromising the recall rates. We also discuss how one other existing popular QSR method named ShoRAH can be improved using this new approach. On benchmark data sets, our estimation method provided accurate richness estimates (< 0.2 median estimation error) and improved the precision of ViQuaS by 2%-13% and F-score by 1%-9% without compromising the recall rates. We also demonstrate that our estimation method can be used to improve the precision and F-score of ShoRAH by 0%-7% and 0%-5% respectively. The proposed probabilistic estimation method can be used to estimate the richness of viral populations with a quasispecies behavior and to improve the accuracy of the quasispecies spectra reconstructed by the existing methods ViQuaS and ShoRAH in the presence of a moderate level of technical sequencing errors. http://sourceforge.net/projects/viquas/

MOTIVATION: Most prokaryotic genomes are circular with a single chromosome (called circular genomes), which consist of bacteria and archaea. Orthologous genes (abbreviated as orthologs) are genes directly evolved from an ancestor gene, and can be traced through different species in evolution. Shared orthologs between bacterial genomes have been used to measure their genome evolution. Here, organization of circular genomes is analyzed via distributions of shared orthologs between genomes. However, these distributions are often asymmetric and bimodal; to date, there is no joint distribution to model such data. This motivated us to develop a family of bivariate distributions with generalized von Mises marginals (BGVM) and its statistical inference. RESULTS: A new measure based on circular grade correlation and the fraction of shared orthologs is proposed for association between circular genomes, and a visualization tool developed to depict genome structure similarity. The proposed procedures are applied to eight pairs of prokaryotes separated from domain down to species, and 13 mycoplasma bacteria that are mammalian pathogens belonging to the same genus. We close with remarks on further applications to many features of genomic organization, e.g. shared transcription factor binding sites, between any pair of circular genomes. Thus, the proposed procedures may be applied to identifying conserved chromosome backbones, among others, for genome construction in synthetic biology. AVAILABILITY: All codes of the BGVM procedures and 1000+ prokaryotic genomes are available at http://www.stat.sinica.edu.tw/∼gshieh/bgvm.htm.

BACKGROUND: In metagenomic studies, a process called binning is necessary to assign contigs that belong to multiple species to their respective phylogenetic groups. Most of the current methods of binning, such as BLAST, k-mer and PhyloPythia, involve assigning sequence fragments by comparing sequence similarity or sequence composition with already-sequenced genomes that are still far from comprehensive. We propose a semi-supervised seeding method for binning that does not depend on knowledge of completed genomes. Instead, it extracts the flanking sequences of highly conserved 16S rRNA from the metagenome and uses them as seeds (labels) to assign other reads based on their compositional similarity. RESULTS: The proposed seeding method is implemented on an unsupervised Growing Self-Organising Map (GSOM), and called Seeded GSOM (S-GSOM). We compared it with four well-known semi-supervised learning methods in a preliminary test, separating random-length prokaryotic sequence fragments sampled from the NCBI genome database. We identified the flanking sequences of the highly conserved 16S rRNA as suitable seeds that could be used to group the sequence fragments according to their species. S-GSOM showed superior performance compared to the semi-supervised methods tested. Additionally, S-GSOM may also be used to visually identify some species that do not have seeds. The proposed method was then applied to simulated metagenomic datasets using two different confidence threshold settings and compared with PhyloPythia, k-mer and BLAST. At the reference taxonomic level Order, S-GSOM outperformed all k-mer and BLAST results and showed comparable results with PhyloPythia for each of the corresponding confidence settings, where S-GSOM performed better than PhyloPythia in the >/= 10 reads datasets and comparable in the > or = 8 kb benchmark tests. CONCLUSION: In the task of binning using semi-supervised learning methods, results indicate S-GSOM to be the best of the methods tested. Most importantly, the proposed method does not require knowledge from known genomes and uses only very few labels (one per species is sufficient in most cases), which are extracted from the metagenome itself. These advantages make it a very attractive binning method. S-GSOM outperformed the binning methods that depend on already-sequenced genomes, and compares well to the current most advanced binning method, PhyloPythia.

BACKGROUND: Existing methods for whole-genome comparisons require prior knowledge of related species and provide little automation in the function prediction process. Bacteriophage genomes are an example that cannot be easily analyzed by these methods. This work addresses these shortcomings and aims to provide an automated prediction system of gene function. RESULTS: We have developed a novel system called SynFPS to perform gene function prediction over completed genomes. The prediction system is initialized by clustering a large collection of weakly related genomes into groups based on their resemblance in gene distribution. From each individual group, data are then extracted and used to train a Support Vector Machine that makes gene function predictions. Experiments were conducted with 9 different gene functions over 296 bacteriophage genomes. Cross validation results gave an average prediction accuracy of ~80%, which is comparable to other genomic-context based prediction methods. Functional predictions are also made on 3 uncharacterized genes and 12 genes that cannot be identified by sequence alignment. The software is publicly available at http://www.synteny.net/. CONCLUSION: The proposed system employs genomic context to predict gene function and detect gene correspondence in whole-genome comparisons. Although our experimental focus is on bacteriophages, the method may be extended to other microbial genomes as they share a number of similar characteristics with phage genomes such as gene order conservation.

MOTIVATION: Current Self-Organizing Maps (SOMs) approaches to gene expression pattern clustering require the user to predefine the number of clusters likely to be expected. Hierarchical clustering methods used in this area do not provide unique partitioning of data. We describe an unsupervised dynamic hierarchical self-organizing approach, which suggests an appropriate number of clusters, to perform class discovery and marker gene identification in microarray data. In the process of class discovery, the proposed algorithm identifies corresponding sets of predictor genes that best distinguish one class from other classes. The approach integrates merits of hierarchical clustering with robustness against noise known from self-organizing approaches. RESULTS: The proposed algorithm applied to DNA microarray data sets of two types of cancers has demonstrated its ability to produce the most suitable number of clusters. Further, the corresponding marker genes identified through the unsupervised algorithm also have a strong biological relationship to the specific cancer class. The algorithm tested on leukemia microarray data, which contains three leukemia types, was able to determine three major and one minor cluster. Prediction models built for the four clusters indicate that the prediction strength for the smaller cluster is generally low, therefore labelled as uncertain cluster. Further analysis shows that the uncertain cluster can be subdivided further, and the subdivisions are related to two of the original clusters. Another test performed using colon cancer microarray data has automatically derived two clusters, which is consistent with the number of classes in data (cancerous and normal). AVAILABILITY: JAVA software of dynamic SOM tree algorithm is available upon request for academic use. SUPPLEMENTARY INFORMATION: A comparison of rectangular and hexagonal topologies for GSOM is available from http://www.mame.mu.oz.au/mechatronics/journalinfo/Hsu2003supp.pdf