Coevolution reveals a network of human proteins originating with multicellularity.

Protein interaction networks play central roles in biological systems, from simple metabolic pathways through complex programs permitting the development of organisms. Multicellularity could only have arisen from a careful orchestration of cellular and molecular roles and responsibilities, all properly controlled and regulated. Disease reflects a breakdown of this organismal homeostasis. To better understand the evolution of interactions whose dysfunction may be contributing factors to disease, we derived the human protein coevolution network using our MatrixMatchMaker algorithm and using the Orthologous MAtrix project (OMA) database as a source for protein orthologs from 103 eukaryotic genomes. We annotated the coevolution network using protein-protein interaction data, many functional data sources, and we explored the evolutionary rates and dates of emergence of the proteins in our data set. Strikingly, clustering based only on the topology of the coevolution network partitions it into two subnetworks, one generally representing ancient eukaryotic functions and the other functions more recently acquired during animal evolution. That latter subnetwork is enriched for proteins with roles in cell-cell communication, the control of cell division, and related multicellular functions. Further annotation using data from genetic disease databases and cancer genome sequences strongly implicates these proteins in both ciliopathies and cancer. The enrichment for such disease markers in the animal network suggests a functional link between these coevolving proteins. Genetic validation corroborates the recruitment of ancient cilia in the evolution of multicellularity.

The human protein coevolution network.

Coevolution maintains interactions between phenotypic traits through the process of reciprocal natural selection. Detecting molecular coevolution can expose functional interactions between molecules in the cell, generating insights into biological processes, pathways, and the networks of interactions important for cellular function. Prediction of interaction partners from different protein families exploits the property that interacting proteins can follow similar patterns and relative rates of evolution. Current methods for detecting coevolution based on the similarity of phylogenetic trees or evolutionary distance matrices have, however, been limited by requiring coevolution over the entire evolutionary history considered and are inaccurate in the presence of paralogous copies. We present a novel method for determining coevolving protein partners by finding the largest common submatrix in a given pair of distance matrices, with the size of the largest common submatrix measuring the strength of coevolution. This approach permits us to consider matrices of different size and scale, to find lineage-specific coevolution, and to predict multiple interaction partners. We used MatrixMatchMaker to predict protein-protein interactions in the human genome. We show that proteins that are known to interact physically are more strongly coevolving than proteins that simply belong to the same biochemical pathway. The human coevolution network is highly connected, suggesting many more protein-protein interactions than are currently known from high-throughput and other experimental evidence. These most strongly coevolving proteins suggest interactions that have been maintained over long periods of evolutionary time, and that are thus likely to be of fundamental importance to cellular function.

Sensitivity and breadth of detection of high-throughput sequencing for adventitious virus detection.

High-throughput sequencing (HTS) is capable of broad virus detection encompassing both known and unknown adventitious viruses in a variety of sample matrices. We describe the development of a general-purpose HTS-based method for the detection of adventitious viruses. Performance was evaluated using 16 viruses equivalent to well-characterized National Institutes of Health (NIH) virus stocks and another six viruses of interest. A viral vaccine crude harvest and a cell substrate matrix were spiked with 22 viruses. Specificity was demonstrated for all 22 viruses at the species level. Our method was capable of detecting and identifying adventitious viruses spiked at 104 genome copies per milliliter in a viral vaccine crude harvest and 0.01 viral genome copies spiked per cell in a cell substrate matrix. Moreover, 9 of the 11 NIH model viruses with published in vivo data were detected by HTS with an equivalent or better sensitivity (in a viral vaccine crude harvest). Our general-purpose HTS method is unbiased and highly sensitive for the detection of adventitious viruses, and has a large breadth of detection, which may obviate the need to perform in vivo testing.

The impact of reticulate evolution on genome phylogeny.

Genome phylogenies are used to build tree-like representations of evolutionary relationships among genomes. However, in condensing the phylogenetic signals within a set of genomes down to a single tree, these methods generally do not explicitly take into account discordant signals arising due to lateral genetic transfer. Because conflicting vertical and horizontal signals can produce compromise trees that do not reflect either type of history, it is essential to understand the sensitivity of inferred genome phylogenies to these confounding effects. Using replicated simulations of genome evolution, we show that different scenarios of lateral genetic transfer have significant impacts on the ability to recover the "true" tree of genomes, even when corrections for phylogenetically discordant signals are used.

Integron-associated gene cassettes in Halifax Harbour: assessment of a mobile gene pool in marine sediments.

The integron/gene cassette systems identified in bacteria comprise a class of genetic elements that allow adaptation by acquisition of gene cassettes. Integron gene cassettes have been shown to facilitate the spread of drug resistance in human pathogens but their role outside a clinical setting has not been explored extensively. We sequenced 2145 integron gene cassettes from four marine sediment samples taken from the vicinity of Halifax Nova Scotia, Canada, increasing the number of gene cassettes obtained from environmental microbial communities by 10-fold. Sequence analyses reveals that the majority of these cassettes encode novel proteins and that this study is consistent with previous claims of high cassette diversity as we estimate a Chao1 diversity index of approximately 3000 cassettes from these samples. The functional distribution of environmental cassettes recovered in this study, when compared with that of cassettes from the only other source with significant sampling (Vibrio genomes) suggests that alternate selection regimes might be acting on these two gene pools. The majority of cassettes recovered in this study encode novel, unknown proteins. In instances where we obtained multiple alleles of a novel protein we demonstrate that non-synonymous versus synonymous substitution rates ratios suggest relaxed selection. Cassette-encoded proteins with known homologues represent a variety of functions and prevalent among these are isochorismatases; proteins involved in iron scavenging. Phylogenetic analysis of these isochorismatases as well as of cassette-encoded acetyltransferases reveals a patchy distribution, suggesting multiple sources for the origin of these cassettes. Finally, the two most environmentally similar sample sites considered in this study display the greatest overlap of cassette types, consistent with the hypothesis that cassette genes encode adaptive proteins.

A simulation test bed for hypotheses of genome evolution.

MOTIVATION: Microbial genomes undergo evolutionary processes such as gene family expansion and contraction, variable rates and patterns of sequence substitution and lateral genetic transfer. Simulation tools are essential for both the generation of data under different evolutionary models and the validation of analytical methods on such data. However, meaningful investigation of phenomena such as lateral genetic transfer requires the simultaneous consideration of many underlying evolutionary processes. RESULTS: We have developed EvolSimulator, a software package that combines non-stationary sequence and gene family evolution together with models of lateral genetic transfer, within a customizable birth-death model of speciation and extinction. Here, we examine simulated data sets generated with EvolSimulator using existing statistical techniques from the evolutionary literature, showing in detail each component of the simulation strategy. AVAILABILITY: Source code, manual and other information are freely available at www.bioinformatics.org.au/evolsim. SUPPLEMENTARY INFORMATION: Supplementary data are available at Bioinformatics online.

Considerations for Optimization of High-Throughput Sequencing Bioinformatics Pipelines for Virus Detection.

High-throughput sequencing (HTS) has demonstrated capabilities for broad virus detection based upon discovery of known and novel viruses in a variety of samples, including clinical, environmental, and biological. An important goal for HTS applications in biologics is to establish parameter settings that can afford adequate sensitivity at an acceptable computational cost (computation time, computer memory, storage, expense or/and efficiency), at critical steps in the bioinformatics pipeline, including initial data quality assessment, trimming/cleaning, and assembly (to reduce data volume and increase likelihood of appropriate sequence identification). Additionally, the quality and reliability of the results depend on the availability of a complete and curated viral database for obtaining accurate results; selection of sequence alignment programs and their configuration, that retains specificity for broad virus detection with reduced false-positive signals; removal of host sequences without loss of endogenous viral sequences of interest; and use of a meaningful reporting format, which can retain critical information of the analysis for presentation of readily interpretable data and actionable results. Furthermore, after alignment, both automated and manual evaluation may be needed to verify the results and help assign a potential risk level to residual, unmapped reads. We hope that the collective considerations discussed in this paper aid toward optimization of data analysis pipelines for virus detection by HTS.

Ancient lateral gene transfer in the evolution of Bdellovibrio bacteriovorus.

The recently sequenced genome of the predatory delta-proteobacterium Bdellovibrio bacteriovorus provides many insights into its metabolism and evolution. Because its genes are reasonably uniform in G+C content, it was suggested that B. bacteriovorus actively resists recombination with foreign DNA and horizontal transfer of DNA from other bacteria. To investigate this further, we carried out a variety of phylogenetic and comparative genomics analyses using data from >200 microbial genomes, including several published delta-proteobacteria. Although there might be little evidence for the extensive recent transfer of genes, we demonstrate that ancient lateral gene acquisition has shaped the B. bacteriovorus genome to a great extent.

A Multicenter Study To Evaluate the Performance of High-Throughput Sequencing for Virus Detection.

The capability of high-throughput sequencing (HTS) for detection of known and unknown viruses makes it a powerful tool for broad microbial investigations, such as evaluation of novel cell substrates that may be used for the development of new biological products. However, like any new assay, regulatory applications of HTS need method standardization. Therefore, our three laboratories initiated a study to evaluate performance of HTS for potential detection of viral adventitious agents by spiking model viruses in different cellular matrices to mimic putative materials for manufacturing of biologics. Four model viruses were selected based upon different physical and biochemical properties and commercial availability: human respiratory syncytial virus (RSV), Epstein-Barr virus (EBV), feline leukemia virus (FeLV), and human reovirus (REO). Additionally, porcine circovirus (PCV) was tested by one laboratory. Independent samples were prepared for HTS by spiking intact viruses or extracted viral nucleic acids, singly or mixed, into different HeLa cell matrices (resuspended whole cells, cell lysate, or total cellular RNA). Data were obtained using different sequencing platforms (Roche 454, Illumina HiSeq1500 or HiSeq2500). Bioinformatic analyses were performed independently by each laboratory using available tools, pipelines, and databases. The results showed that comparable virus detection was obtained in the three laboratories regardless of sample processing, library preparation, sequencing platform, and bioinformatic analysis: between 0.1 and 3 viral genome copies per cell were detected for all of the model viruses used. This study highlights the potential for using HTS for sensitive detection of adventitious viruses in complex biological samples containing cellular background. IMPORTANCE Recent high-throughput sequencing (HTS) investigations have resulted in unexpected discoveries of known and novel viruses in a variety of sample types, including research materials, clinical materials, and biological products. Therefore, HTS can be a powerful tool for supplementing current methods for demonstrating the absence of adventitious or unwanted viruses in biological products, particularly when using a new cell line. However, HTS is a complex technology with different platforms, which needs standardization for evaluation of biologics. This collaborative study was undertaken to investigate detection of different virus types using two different HTS platforms. The results of the independently performed studies demonstrated a similar sensitivity of virus detection, regardless of the different sample preparation and processing procedures and bioinformatic analyses done in the three laboratories. Comparable HTS detection of different virus types supports future development of reference virus materials for standardization and validation of different HTS platforms.

Advanced Virus Detection Technologies Interest Group (AVDTIG): Efforts on High Throughput Sequencing (HTS) for Virus Detection.

Several nucleic-acid based technologies have recently emerged with capabilities for broad virus detection. One of these, high throughput sequencing, has the potential for novel virus detection because this method does not depend upon prior viral sequence knowledge. However, the use of high throughput sequencing for testing biologicals poses greater challenges as compared to other newly introduced tests due to its technical complexities and big data bioinformatics. Thus, the Advanced Virus Detection Technologies Users Group was formed as a joint effort by regulatory and industry scientists to facilitate discussions and provide a forum for sharing data and experiences using advanced new virus detection technologies, with a focus on high throughput sequencing technologies. The group was initiated as a task force that was coordinated by the Parenteral Drug Association and subsequently became the Advanced Virus Detection Technologies Interest Group to continue efforts for using new technologies for detection of adventitious viruses with broader participation, including international government agencies, academia, and technology service providers.

GANN: genetic algorithm neural networks for the detection of conserved combinations of features in DNA.

BACKGROUND: The multitude of motif detection algorithms developed to date have largely focused on the detection of patterns in primary sequence. Since sequence-dependent DNA structure and flexibility may also play a role in protein-DNA interactions, the simultaneous exploration of sequence- and structure-based hypotheses about the composition of binding sites and the ordering of features in a regulatory region should be considered as well. The consideration of structural features requires the development of new detection tools that can deal with data types other than primary sequence. RESULTS: GANN (available at http://bioinformatics.org.au/gann) is a machine learning tool for the detection of conserved features in DNA. The software suite contains programs to extract different regions of genomic DNA from flat files and convert these sequences to indices that reflect sequence and structural composition or the presence of specific protein binding sites. The machine learning component allows the classification of different types of sequences based on subsamples of these indices, and can identify the best combinations of indices and machine learning architecture for sequence discrimination. Another key feature of GANN is the replicated splitting of data into training and test sets, and the implementation of negative controls. In validation experiments, GANN successfully merged important sequence and structural features to yield good predictive models for synthetic and real regulatory regions. CONCLUSION: GANN is a flexible tool that can search through large sets of sequence and structural feature combinations to identify those that best characterize a set of sequences.

Deduction of probable events of lateral gene transfer through comparison of phylogenetic trees by recursive consolidation and rearrangement.

BACKGROUND: When organismal phylogenies based on sequences of single marker genes are poorly resolved, a logical approach is to add more markers, on the assumption that weak but congruent phylogenetic signal will be reinforced in such multigene trees. Such approaches are valid only when the several markers indeed have identical phylogenies, an issue which many multigene methods (such as the use of concatenated gene sequences or the assembly of supertrees) do not directly address. Indeed, even when the true history is a mixture of vertical descent for some genes and lateral gene transfer (LGT) for others, such methods produce unique topologies. RESULTS: We have developed software that aims to extract evidence for vertical and lateral inheritance from a set of gene trees compared against an arbitrary reference tree. This evidence is then displayed as a synthesis showing support over the tree for vertical inheritance, overlaid with explicit lateral gene transfer (LGT) events inferred to have occurred over the history of the tree. Like splits-tree methods, one can thus identify nodes at which conflict occurs. Additionally one can make reasonable inferences about vertical and lateral signal, assigning putative donors and recipients. CONCLUSION: A tool such as ours can serve to explore the reticulated dimensionality of molecular evolution, by dissecting vertical and lateral inheritance at high resolution. By this, we mean that individual nodes can be examined not only for congruence, but also for coherence in light of LGT. We assert that our tools will facilitate the comparison of phylogenetic trees, and the interpretation of conflicting data.

Characterization of species-specific genes using a flexible, web-based querying system.

We describe a query-based web-accessible system (www.neurogadgets.com/bws.php) for facilitating comparative microbial genomics. A variety of query pages are available, each with numerous options, that allow a biologist to pose relevant questions of genomic data. We illustrate with a characterization of species-specific protein-coding genes (so-called "ORFans"), finding that they are on average smaller, faster evolving, and less G+C-rich, and that they encode proteins more basic in their predicted isoelectric point, compared with non-species-specific genes. Using a dual-threshold approach, we conclude that these are characteristics of true species-specific genes, rather than artifacts of mis-annotation.

Cataloguing the taxonomic origins of sequences from a heterogeneous sample using phylogenomics: applications in adventitious agent detection.

We have designed and implemented a software system, named PhyloID™, that can be used to detect putative adventitious agents in biological samples characterized by next-generation sequencing. PhyloID is run in two steps, each being a self-contained automated process amenable to GMP validation. The first module, MiLY, is responsible for assembling individual sequence reads into contigs, and annotating all sequences with a unique sequence identifier, the number of reads in each contig, and the length of the sequence. The trimmed, assembled and annotated data are then processed by PhyloID's second module, NGmapper. NGmapper takes the FASTA-formatted output from MiLY and identifies the taxonomic origins of the contigs and singletons therein. It compares each sequence's BLASTN hit profile against the patterns of evolutionary relationships described within phylogenomic distance matrices for all of the various taxonomic groups, in order to find the best fit. NGmapper then produces lists of taxonomic assignments in both summarized and detailed form, and tree files for viewing results graphically. We optimized PhyloID's parameters and measured its performance using simulated metagenomic data and subsets of the reference phylogenies. PhyloID's precision and recall in identifying simulated sequences were measured by information retrieval analysis, focusing on read length, read number, sequence accuracy, background complexity, taxonomy and reference data coverage. We found PhyloID to be highly accurate and quantitative in its taxonomic mapping of sequences, with excellent precision, sensitivity and robustness. The degree of taxonomic representation available in publicly available databases remains an issue, as expected, for any sequence classifier, but community sequencing efforts are poised to overcome this problem. In order to illustrate real-world usage of the application, we also describe some simple spike-recovery experiments as well as a multi-site comparative characterization of a viral suspension. These data help to illustrate, to corroborate, and to extend results using simulated data. LAY ABSTRACT: In order to address gaps in the detection of contaminating viruses and microorganisms in vaccines and other biologicals, manufacturers are exploring the use of new technologies that promise greater sensitivity and breadth of coverage. One challenge in implementing such new methods is the complexity of analysis of the "big data" generated by these new instruments: hundreds of millions of sequence reads (segments of genetic material from viruses and cells) need to be compared against a vast and growing number of entries in genetic databases, in order to come up with a confident identification. These large-scale analyses must furthermore be carried out within the strict regulatory environment that governs the industry. We have developed an automated software pipeline named PhyloID™ that is capable of identifying viruses and microorganisms from large-scale sequence data. Using simulated data as well as real samples, we show that PhyloID is both sensitive and accurate in identifying any type of potential contaminant. Such a powerful new assay will be an important addition to the adventitious agent testing package, providing further assurance about product safety.

Preliminary Evaluation of Next-Generation Sequencing Performance Relative to qPCR and In Vitro Cell Culture Tests for Human Cytomegalovirus.

To compare the performances of conventional in vitro indicator cell culture, quantitative polymerase chain reaction (qPCR), and next-generation sequencing (NGS) as detection methods for adventitious agents, a preliminary assessment was performed using human cytomegalovirus (HCMV) as a model virus. HCMV was spiked into a crude viral harvest at various concentrations and inoculated onto MRC-5 cell monolayers. The cultures were observed for cytopathic effects (CPEs) as per the compendial method requirements, and samples were taken at various time points for analysis by qPCR or NGS. When using NGS, the detection of HCMV is 10 fold more sensitive than observed using the conventional CPE endpoint method. It may be possible for qPCR to achieve the detection level demonstrated by NGS, but further optimization of the technique would be required. In addition, NGS was able to detect HCMV in the initial inoculum when it was spiked in at 10 CCID50/mL, suggesting the potential to eliminate cell culture amplification with an NGS-based assay. This study highlights the advantage of NGS as a surrogate broad-spectrum technology for the detection of adventitious agents in biologics. LAY ABSTRACT: Human cytomeglovirus (HCMV) is highly prevalent in the general population and can lead to serious health issues in both immumocompromised individuals and/or newborns. The testing of HCMV in biological materials is stipulated by multiple regulatory agencies where HCMV is a potential risk. This test involves inoculating cell lines that are susceptible to HCMV infection, incubating the cultures for 28 days, and observing the cells for signs of viral infection. Next-generation sequencing and quantitative polymerase chain reaction (qPCR) are two technologies that can potentially shorten the extended 28 day cell culture incubation. In this study, we compared the sensitivity of the compendial cell culture assay with NGS and qPCR for the detection of HCMV. Our results show that NGS can potentially achieve a detection limit that is 10 times more sensitive than the cell culture assay. In addition, NGS was able to detect HCMV in the initial inoculum, potentially eliminating the need for cell culture amplification of the virus. Finally, sequence data generated by NGS directly demonstrate the presence of HCMV, and such information can serve as the foundation for any follow-up investigation.

Using coevolution to predict protein-protein interactions.

Bioinformatic methods to predict protein-protein interactions (PPI) via coevolutionary analysis have -positioned themselves to compete alongside established in vitro methods, despite a lack of understanding for the underlying molecular mechanisms of the coevolutionary process. Investigating the alignment of coevolutionary predictions of PPI with experimental data can focus the effective scope of prediction and lead to better accuracies. A new rate-based coevolutionary method, MMM, preferentially finds obligate interacting proteins that form complexes, conforming to results from studies based on coimmunoprecipitation coupled with mass spectrometry. Using gold-standard databases as a benchmark for accuracy, MMM surpasses methods based on abundance ratios, suggesting that correlated evolutionary rates may yet be better than coexpression at predicting interacting proteins. At the level of protein domains, -coevolution is difficult to detect, even with MMM, except when considering small-scale experimental data involving proteins with multiple domains. Overall, these findings confirm that coevolutionary -methods can be confidently used in predicting PPI, either independently or as drivers of coimmunoprecipitation experiments.

Phylogenetic analyses of cyanobacterial genomes: quantification of horizontal gene transfer events.

Using 1128 protein-coding gene families from 11 completely sequenced cyanobacterial genomes, we attempt to quantify horizontal gene transfer events within cyanobacteria, as well as between cyanobacteria and other phyla. A novel method of detecting and enumerating potential horizontal gene transfer events within a group of organisms based on analyses of "embedded quartets" allows us to identify phylogenetic signal consistent with a plurality of gene families, as well as to delineate cases of conflict to the plurality signal, which include horizontally transferred genes. To infer horizontal gene transfer events between cyanobacteria and other phyla, we added homologs from 168 available genomes. We screened phylogenetic trees reconstructed for each of these extended gene families for highly supported monophyly of cyanobacteria (or lack of it). Cyanobacterial genomes reveal a complex evolutionary history, which cannot be represented by a single strictly bifurcating tree for all genes or even most genes, although a single completely resolved phylogeny was recovered from the quartets' plurality signals. We find more conflicts within cyanobacteria than between cyanobacteria and other phyla. We also find that genes from all functional categories are subject to transfer. However, in interphylum as compared to intraphylum transfers, the proportion of metabolic (operational) gene transfers increases, while the proportion of informational gene transfers decreases.

Weighted genome trees: refinements and applications.

There are many ways to group completed genome sequences in hierarchical patterns (trees) reflecting relationships between their genes. Such groupings help us organize biological information and bear crucially on underlying processes of genome and organismal evolution. Genome trees make use of all comparable genes but can variously weight the contributions of these genes according to similarity, congruent patterns of similarity, or prevalence among genomes. Here we explore such possible weighting strategies, in an analysis of 142 prokaryotic and 5 eukaryotic genomes. We demonstrate that alternate weighting strategies have different advantages, and we propose that each may have its specific uses in systematic or evolutionary biology. Comparisons of results obtained with different methods can provide further clues to major events and processes in genome evolution.