Gene identification and protein classification in microbial metagenomic sequence data via incremental clustering

Background  

The identification and study of proteins from metagenomic datasets can shed light on the roles and interactions of the source organisms in their communities. However, metagenomic datasets are characterized by the presence of organisms with varying GC composition, codon usage biases etc., and consequently gene identification is challenging. The vast amount of sequence data also requires faster protein family classification tools.     

Taxonomic classification of metagenomic shotgun sequences with CARMA3

The vast majority of microbes are unculturable and thus cannot be sequenced by means of traditional methods. High-throughput sequencing techniques like 454 or Solexa-Illumina make it possible to explore those microbes by studying whole natural microbial communities and analysing their biological diversity as well as the underlying metabolic pathways. Over the past few years, different methods have been developed for the taxonomic and functional characterization of metagenomic shotgun sequences. However, the taxonomic classification of metagenomic sequences from novel species without close homologue in the biological sequence databases poses a challenge due to the high number of wrong taxonomic predictions on lower taxonomic ranks. Here we present CARMA3, a new method for the taxonomic classification of assembled and unassembled metagenomic sequences that has been adapted to work with both BLAST and HMMER3 homology searches. We show that our method makes fewer wrong taxonomic predictions (at the same sensitivity) than other BLAST-based methods. CARMA3 is freely accessible via the web application WebCARMA from http://webcarma.cebitec.uni-bielefeld.de.     

TWARIT: an extremely rapid and efficient approach for phylogenetic classification of metagenomic sequences

Phylogenetic assignment of individual sequence reads to their respective taxa, referred to as 'taxonomic binning', constitutes a key step of metagenomic analysis. Existing binning methods have limitations either with respect to time or accuracy/specificity of binning. Given these limitations, development of a method that can bin vast amounts of metagenomic sequence data in a rapid, efficient and computationally inexpensive manner can profoundly influence metagenomic analysis in computational resource poor settings. We introduce TWARIT, a hybrid binning algorithm, that employs a combination of short-read alignment and composition-based signature sorting approaches to achieve rapid binning rates without compromising on binning accuracy and specificity. TWARIT is validated with simulated and real-world metagenomes and the results demonstrate significantly lower overall binning times compared to that of existing methods. Furthermore, the binning accuracy and specificity of TWARIT are observed to be comparable/superior to them. A web server implementing TWARIT algorithm is available at http://metagenomics.atc.tcs.com/Twarit/     

Clustering metagenomic sequences with interpolated Markov models

Background  

Sequencing of environmental DNA (often called metagenomics) has shown tremendous potential to uncover the vast number of unknown microbes that cannot be cultured and sequenced by traditional methods. Because the output from metagenomic sequencing is a large set of reads of unknown origin, clustering reads together that were sequenced from the same species is a crucial analysis step. Many effective approaches to this task rely on sequenced genomes in public databases, but these genomes are a highly biased sample that is not necessarily representative of environments interesting to many metagenomics projects.     

A framework for space-efficient read clustering in metagenomic samples

Background

A metagenomic sample is a set of DNA fragments, randomly extracted from multiple cells in an environment, belonging to distinct, often unknown species. Unsupervised metagenomic clustering aims at partitioning a metagenomic sample into sets that approximate taxonomic units, without using reference genomes. Since samples are large and steadily growing, space-efficient clustering algorithms are strongly needed.

Results

We design and implement a space-efficient algorithmic framework that solves a number of core primitives in unsupervised metagenomic clustering using just the bidirectional Burrows-Wheeler index and a union-find data structure on the set of reads. When run on a sample of total length n, with m reads of maximum length ? each, on an alphabet of total size σ, our algorithms take O(n(t+logσ)) time and just 2n+o(n)+O(max{? σlogn,K logm}) bits of space in addition to the index and to the union-find data structure, where K is a measure of the redundancy of the sample and t is the query time of the union-find data structure.

Conclusions

Our experimental results show that our algorithms are practical, they can exploit multiple cores by a parallel traversal of the suffix-link tree, and they are competitive both in space and in time with the state of the art.

     

Composition-based classification of short metagenomic sequences elucidates the landscapes of taxonomic and functional enrichment of microorganisms

Compared with traditional algorithms for long metagenomic sequence classification, characterizing microorganisms’ taxonomic and functional abundance based on tens of millions of very short reads are much more challenging. We describe an efficient composition and phylogeny-based algorithm [Metagenome Composition Vector (MetaCV)] to classify very short metagenomic reads (75–100 bp) into specific taxonomic and functional groups. We applied MetaCV to the Meta-HIT data (371-Gb 75-bp reads of 109 human gut metagenomes), and this single-read-based, instead of assembly-based, classification has a high resolution to characterize the composition and structure of human gut microbiota, especially for low abundance species. Most strikingly, it only took MetaCV 10 days to do all the computation work on a server with five 24-core nodes. To our knowledge, MetaCV, benefited from the strategy of composition comparison, is the first algorithm that can classify millions of very short reads within affordable time.     

Tumor classification and marker gene prediction by feature selection and fuzzy c-means clustering using microarray data

Background  

Using DNA microarrays, we have developed two novel models for tumor classification and target gene prediction. First, gene expression profiles are summarized by optimally selected Self-Organizing Maps (SOMs), followed by tumor sample classification by Fuzzy C-means clustering. Then, the prediction of marker genes is accomplished by either manual feature selection (visualizing the weighted/mean SOM component plane) or automatic feature selection (by pair-wise Fisher's linear discriminant).     

Genomic-enabled prediction with classification algorithms

Pearson''s correlation coefficient (ρ) is the most commonly reported metric of the success of prediction in genomic selection (GS). However, in real breeding ρ may not be very useful for assessing the quality of the regression in the tails of the distribution, where individuals are chosen for selection. This research used 14 maize and 16 wheat data sets with different trait–environment combinations. Six different models were evaluated by means of a cross-validation scheme (50 random partitions each, with 90% of the individuals in the training set and 10% in the testing set). The predictive accuracy of these algorithms for selecting individuals belonging to the best α=10, 15, 20, 25, 30, 35, 40% of the distribution was estimated using Cohen''s kappa coefficient (κ) and an ad hoc measure, which we call relative efficiency (RE), which indicates the expected genetic gain due to selection when individuals are selected based on GS exclusively. We put special emphasis on the analysis for α=15%, because it is a percentile commonly used in plant breeding programmes (for example, at CIMMYT). We also used ρ as a criterion for overall success. The algorithms used were: Bayesian LASSO (BL), Ridge Regression (RR), Reproducing Kernel Hilbert Spaces (RHKS), Random Forest Regression (RFR), and Support Vector Regression (SVR) with linear (lin) and Gaussian kernels (rbf). The performance of regression methods for selecting the best individuals was compared with that of three supervised classification algorithms: Random Forest Classification (RFC) and Support Vector Classification (SVC) with linear (lin) and Gaussian (rbf) kernels. Classification methods were evaluated using the same cross-validation scheme but with the response vector of the original training sets dichotomised using a given threshold. For α=15%, SVC-lin presented the highest κ coefficients in 13 of the 14 maize data sets, with best values ranging from 0.131 to 0.722 (statistically significant in 9 data sets) and the best RE in the same 13 data sets, with values ranging from 0.393 to 0.948 (statistically significant in 12 data sets). RR produced the best mean for both κ and RE in one data set (0.148 and 0.381, respectively). Regarding the wheat data sets, SVC-lin presented the best κ in 12 of the 16 data sets, with outcomes ranging from 0.280 to 0.580 (statistically significant in 4 data sets) and the best RE in 9 data sets ranging from 0.484 to 0.821 (statistically significant in 5 data sets). SVC-rbf (0.235), RR (0.265) and RHKS (0.422) gave the best κ in one data set each, while RHKS and BL tied for the last one (0.234). Finally, BL presented the best RE in two data sets (0.738 and 0.750), RFR (0.636) and SVC-rbf (0.617) in one and RHKS in the remaining three (0.502, 0.458 and 0.586). The difference between the performance of SVC-lin and that of the rest of the models was not so pronounced at higher percentiles of the distribution. The behaviour of regression and classification algorithms varied markedly when selection was done at different thresholds, that is, κ and RE for each algorithm depended strongly on the selection percentile. Based on the results, we propose classification method as a promising alternative for GS in plant breeding.     

Fast and accurate taxonomic assignments of metagenomic sequences using MetaBin

Taxonomic assignment of sequence reads is a challenging task in metagenomic data analysis, for which the present methods mainly use either composition- or homology-based approaches. Though the homology-based methods are more sensitive and accurate, they suffer primarily due to the time needed to generate the Blast alignments. We developed the MetaBin program and web server for better homology-based taxonomic assignments using an ORF-based approach. By implementing Blat as the faster alignment method in place of Blastx, the analysis time has been reduced by severalfold. It is benchmarked using both simulated and real metagenomic datasets, and can be used for both single and paired-end sequence reads of varying lengths (≥45 bp). To our knowledge, MetaBin is the only available program that can be used for the taxonomic binning of short reads (<100 bp) with high accuracy and high sensitivity using a homology-based approach. The MetaBin web server can be used to carry out the taxonomic analysis, by either submitting reads or Blastx output. It provides several options including construction of taxonomic trees, creation of a composition chart, functional analysis using COGs, and comparative analysis of multiple metagenomic datasets. MetaBin web server and a standalone version for high-throughput analysis are available freely at http://metabin.riken.jp/.     

Gene prioritization and clustering by multi-view text mining

Background  

Text mining has become a useful tool for biologists trying to understand the genetics of diseases. In particular, it can help identify the most interesting candidate genes for a disease for further experimental analysis. Many text mining approaches have been introduced, but the effect of disease-gene identification varies in different text mining models. Thus, the idea of incorporating more text mining models may be beneficial to obtain more refined and accurate knowledge. However, how to effectively combine these models still remains a challenging question in machine learning. In particular, it is a non-trivial issue to guarantee that the integrated model performs better than the best individual model.     

Gene structure prediction by spliced alignment of genomic DNA with protein sequences: increased accuracy by differential splice site scoring

Gene identification in genomic DNA from eukaryotes is complicated by the vast combinatorial possibilities of potential exon assemblies. If the gene encodes a protein that is closely related to known proteins, gene identification is aided by matching similarity of potential translation products to those target proteins. The genomic DNA and protein sequences can be aligned directly by scoring the implied residues of in-frame nucleotide triplets against the protein residues in conventional ways, while allowing for long gaps in the alignment corresponding to introns in the genomic DNA. We describe a novel method for such spliced alignment. The method derives an optimal alignment based on scoring for both sequence similarity of the predicted gene product to the protein sequence and intrinsic splice site strength of the predicted introns. Application of the method to a representative set of 50 known genes from Arabidopsis thaliana showed significant improvement in prediction accuracy compared to previous spliced alignment methods. The method is also more accurate than ab initio gene prediction methods, provided sufficiently close target proteins are available. In view of the fast growth of public sequence repositories, we argue that close targets will be available for the majority of novel genes, making spliced alignment an excellent practical tool for high-throughput automated genome annotation.     

Prediction and identification of sequences coding for orphan enzymes using genomic and metagenomic neighbours

Despite the current wealth of sequencing data, one‐third of all biochemically characterized metabolic enzymes lack a corresponding gene or protein sequence, and as such can be considered orphan enzymes. They represent a major gap between our molecular and biochemical knowledge, and consequently are not amenable to modern systemic analyses. As 555 of these orphan enzymes have metabolic pathway neighbours, we developed a global framework that utilizes the pathway and (meta)genomic neighbour information to assign candidate sequences to orphan enzymes. For 131 orphan enzymes (37% of those for which (meta)genomic neighbours are available), we associate sequences to them using scoring parameters with an estimated accuracy of 70%, implying functional annotation of 16 345 gene sequences in numerous (meta)genomes. As a case in point, two of these candidate sequences were experimentally validated to encode the predicted activity. In addition, we augmented the currently available genome‐scale metabolic models with these new sequence–function associations and were able to expand the models by on average 8%, with a considerable change in the flux connectivity patterns and improved essentiality prediction.     

Eu-Detect: An algorithm for detecting eukaryotic sequences in metagenomic data sets

Physical partitioning techniques are routinely employed (during sample preparation stage) for segregating the prokaryotic and eukaryotic fractions of metagenomic samples. In spite of these efforts, several metagenomic studies focusing on bacterial and archaeal populations have reported the presence of contaminating eukaryotic sequences in metagenomic data sets. Contaminating sequences originate not only from genomes of micro-eukaryotic species but also from genomes of (higher) eukaryotic host cells. The latter scenario usually occurs in the case of host-associated metagenomes. Identification and removal of contaminating sequences is important, since these sequences not only impact estimates of microbial diversity but also affect the accuracy of several downstream analyses. Currently, the computational techniques used for identifying contaminating eukaryotic sequences, being alignment based, are slow, inefficient, and require huge computing resources. In this article, we present Eu-Detect, an alignment-free algorithm that can rapidly identify eukaryotic sequences contaminating metagenomic data sets. Validation results indicate that on a desktop with modest hardware specifications, the Eu-Detect algorithm is able to rapidly segregate DNA sequence fragments of prokaryotic and eukaryotic origin, with high sensitivity. A Web server for the Eu-Detect algorithm is available at http://metagenomics.atc.tcs.com/Eu-Detect/.     

MetaBinG2: a fast and accurate metagenomic sequence classification system for samples with many unknown organisms

Background

Many methods have been developed for metagenomic sequence classification, and most of them depend heavily on genome sequences of the known organisms. A large portion of sequencing sequences may be classified as unknown, which greatly impairs our understanding of the whole sample.

Result

Here we present MetaBinG2, a fast method for metagenomic sequence classification, especially for samples with a large number of unknown organisms. MetaBinG2 is based on sequence composition, and uses GPUs to accelerate its speed. A million 100 bp Illumina sequences can be classified in about 1 min on a computer with one GPU card. We evaluated MetaBinG2 by comparing it to multiple popular existing methods. We then applied MetaBinG2 to the dataset of MetaSUB Inter-City Challenge provided by CAMDA data analysis contest and compared community composition structures for environmental samples from different public places across cities.

Conclusion

Compared to existing methods, MetaBinG2 is fast and accurate, especially for those samples with significant proportions of unknown organisms.

Reviewers

This article was reviewed by Drs. Eran Elhaik, Nicolas Rascovan, and Serghei Mangul.