Microbial phenomics information extractor (MicroPIE): a natural language processing tool for the automated acquisition of prokaryotic phenotypic characters from text sources.

BACKGROUND: The large-scale analysis of phenomic data (i.e., full phenotypic traits of an organism, such as shape, metabolic substrates, and growth conditions) in microbial bioinformatics has been hampered by the lack of tools to rapidly and accurately extract phenotypic data from existing legacy text in the field of microbiology. To quickly obtain knowledge on the distribution and evolution of microbial traits, an information extraction system needed to be developed to extract phenotypic characters from large numbers of taxonomic descriptions so they can be used as input to existing phylogenetic analysis software packages. RESULTS: We report the development and evaluation of Microbial Phenomics Information Extractor (MicroPIE, version 0.1.0). MicroPIE is a natural language processing application that uses a robust supervised classification algorithm (Support Vector Machine) to identify characters from sentences in prokaryotic taxonomic descriptions, followed by a combination of algorithms applying linguistic rules with groups of known terms to extract characters as well as character states. The input to MicroPIE is a set of taxonomic descriptions (clean text). The output is a taxon-by-character matrix-with taxa in the rows and a set of 42 pre-defined characters (e.g., optimum growth temperature) in the columns. The performance of MicroPIE was evaluated against a gold standard matrix and another student-made matrix. Results show that, compared to the gold standard, MicroPIE extracted 21 characters (50%) with a Relaxed F1 score > 0.80 and 16 characters (38%) with Relaxed F1 scores ranging between 0.50 and 0.80. Inclusion of a character prediction component (SVM) improved the overall performance of MicroPIE, notably the precision. Evaluated against the same gold standard, MicroPIE performed significantly better than the undergraduate students. CONCLUSION: MicroPIE is a promising new tool for the rapid and efficient extraction of phenotypic character information from prokaryotic taxonomic descriptions. However, further development, including incorporation of ontologies, will be necessary to improve the performance of the extraction for some character types.

A Comprehensive Study of Cyanobacterial Morphological and Ecological Evolutionary Dynamics through Deep Geologic Time.

Cyanobacteria have exerted a profound influence on the progressive oxygenation of Earth. As a complementary approach to examining the geologic record-phylogenomic and trait evolutionary analyses of extant species can lead to new insights. We constructed new phylogenomic trees and analyzed phenotypic trait data using novel phylogenetic comparative methods. We elucidated the dynamics of trait evolution in Cyanobacteria over billion-year timescales, and provide evidence that major geologic events in early Earth's history have shaped-and been shaped by-evolution in Cyanobacteria. We identify a robust core cyanobacterial phylogeny and a smaller set of taxa that exhibit long-branch attraction artifacts. We estimated the age of nodes and reconstruct the ancestral character states of 43 phenotypic characters. We find high levels of phylogenetic signal for nearly all traits, indicating the phylogeny carries substantial predictive power. The earliest cyanobacterial lineages likely lived in freshwater habitats, had small cell diameters, were benthic or sessile, and possibly epilithic/endolithic with a sheath. We jointly analyzed a subset of 25 binary traits to determine whether rates of trait evolution have shifted over time in conjunction with major geologic events. Phylogenetic comparative analysis reveal an overriding signal of decreasing rates of trait evolution through time. Furthermore, the data suggest two major rate shifts in trait evolution associated with bursts of evolutionary innovation. The first rate shift occurs in the aftermath of the Great Oxidation Event and "Snowball Earth" glaciations and is associated with decrease in the evolutionary rates around 1.8-1.6 Ga. This rate shift seems to indicate the end of a major diversification of cyanobacterial phenotypes-particularly related to traits associated with filamentous morphology, heterocysts and motility in freshwater ecosystems. Another burst appears around the time of the Neoproterozoic Oxidation Event in the Neoproterozoic, and is associated with the acquisition of traits involved in planktonic growth in marine habitats. Our results demonstrate how uniting genomic and phenotypic datasets in extant bacterial species can shed light on billion-year old events in Earth's history.

MicrO: an ontology of phenotypic and metabolic characters, assays, and culture media found in prokaryotic taxonomic descriptions.

BACKGROUND: MicrO is an ontology of microbiological terms, including prokaryotic qualities and processes, material entities (such as cell components), chemical entities (such as microbiological culture media and medium ingredients), and assays. The ontology was built to support the ongoing development of a natural language processing algorithm, MicroPIE (or, Microbial Phenomics Information Extractor). During the MicroPIE design process, we realized there was a need for a prokaryotic ontology which would capture the evolutionary diversity of phenotypes and metabolic processes across the tree of life, capture the diversity of synonyms and information contained in the taxonomic literature, and relate microbiological entities and processes to terms in a large number of other ontologies, most particularly the Gene Ontology (GO), the Phenotypic Quality Ontology (PATO), and the Chemical Entities of Biological Interest (ChEBI). We thus constructed MicrO to be rich in logical axioms and synonyms gathered from the taxonomic literature. RESULTS: MicrO currently has ~14550 classes (~2550 of which are new, the remainder being microbiologically-relevant classes imported from other ontologies), connected by ~24,130 logical axioms (5,446 of which are new), and is available at (http://purl.obolibrary.org/obo/MicrO.owl) and on the project website at https://github.com/carrineblank/MicrO. MicrO has been integrated into the OBO Foundry Library (http://www.obofoundry.org/ontology/micro.html), so that other ontologies can borrow and re-use classes. Term requests and user feedback can be made using MicrO's Issue Tracker in GitHub. We designed MicrO such that it can support the ongoing and future development of algorithms that can leverage the controlled vocabulary and logical inference power provided by the ontology. CONCLUSIONS: By connecting microbial classes with large numbers of chemical entities, material entities, biological processes, molecular functions, and qualities using a dense array of logical axioms, we intend MicrO to be a powerful new tool to increase the computing power of bioinformatics tools such as the automated text mining of prokaryotic taxonomic descriptions using natural language processing. We also intend MicrO to support the development of new bioinformatics tools that aim to develop new connections between microbial phenotypes and genotypes (i.e., the gene content in genomes). Future ontology development will include incorporation of pathogenic phenotypes and prokaryotic habitats.

Next-generation phenomics for the Tree of Life.

The phenotype represents a critical interface between the genome and the environment in which organisms live and evolve. Phenotypic characters also are a rich source of biodiversity data for tree building, and they enable scientists to reconstruct the evolutionary history of organisms, including most fossil taxa, for which genetic data are unavailable. Therefore, phenotypic data are necessary for building a comprehensive Tree of Life. In contrast to recent advances in molecular sequencing, which has become faster and cheaper through recent technological advances, phenotypic data collection remains often prohibitively slow and expensive. The next-generation phenomics project is a collaborative, multidisciplinary effort to leverage advances in image analysis, crowdsourcing, and natural language processing to develop and implement novel approaches for discovering and scoring the phenome, the collection of phentotypic characters for a species. This research represents a new approach to data collection that has the potential to transform phylogenetics research and to enable rapid advances in constructing the Tree of Life. Our goal is to assemble large phenomic datasets built using new methods and to provide the public and scientific community with tools for phenomic data assembly that will enable rapid and automated study of phenotypes across the Tree of Life.

Phylogenetic distribution of compatible solute synthesis genes support a freshwater origin for cyanobacteria.

Previous work using ancestral state reconstruction of habitat salinity preference proposed that the early cyanobacteria likely lived in a freshwater environment. The aim of this study was to test that hypothesis by performing phylogenetic analyses of the genes underlying salinity preferences in the cyanobacteria. Phylogenetic analysis of compatible solute genes shows that sucrose synthesis genes were likely ancestral in the cyanobacteria, and were also likely inherited during the cyanobacterial endosymbiosis and into the photosynthetic algae and land plants. In addition, the genes for the synthesis of compatible solutes that are necessary for survival in marine and hypersaline environments (such as glucosylglycerol, glucosylglycerate, and glycine betaine) were likely acquired independently high up (i.e., more recently) in the cyanobacterial tree. Because sucrose synthesis is strongly associated with growth in a low salinity environment, this independently supports a freshwater origin for the cyanobacteria. It is also consistent with geologic evidence showing that the early oceans were much warmer and saltier than modern oceans-sucrose synthesis alone would have been insufficient for early cyanobacteria to have colonized early Precambrian oceans that had a higher ionic strength. Indeed, the acquisition of an expanded set of new compatible solute genes may have enabled the historical colonization of marine and hypersaline environments by cyanobacteria, midway through their evolutionary history.

Origin and early evolution of photosynthetic eukaryotes in freshwater environments: reinterpreting proterozoic paleobiology and biogeochemical processes in light of trait evolution.

Phylogenetic analyses were performed on concatenated data sets of 31 genes and 11,789 unambiguously alignable characters from 37 cyanobacterial and 35 chloroplast genomes. The plastid lineage emerged somewhat early in the cyanobacterial tree, at a time when Cyanobacteria were likely unicellular and restricted to freshwater ecosystems. Using relaxed molecular clocks and 22 age constraints spanning cyanobacterial and eukaryote nodes, the common ancestor to the photosynthetic eukaryotes was predicted to have also inhabited freshwater environments around the time that oxygen appeared in the atmosphere (2.0-2.3 Ga). Early diversifications within each of the three major plastid clades were also inferred to have occurred in freshwater environments, through the late Paleoproterozoic and into the middle Mesoproterozoic. The colonization of marine environments by photosynthetic eukaryotes may not have occurred until after the middle Mesoproterozoic (1.2-1.5 Ga). The evolutionary hypotheses proposed here predict that early photosynthetic eukaryotes may have never experienced the widespread anoxia or euxinia suggested to have characterized marine environments in the Paleoproterozoic to early Mesoproterozoic. It also proposes that earliest acritarchs (1.5-1.7 Ga) may have been produced by freshwater taxa. This study highlights how the early evolution of habitat preference in photosynthetic eukaryotes, along with Cyanobacteria, could have contributed to changing biogeochemical conditions on the early Earth.

Low rates of lateral gene transfer among metabolic genes define the evolving biogeochemical niches of archaea through deep time.

Phylogenomic analyses of archaeal genome sequences are providing windows into the group's evolutionary past, even though most archaeal taxa lack a conventional fossil record. Here, phylogenetic analyses were performed using key metabolic genes that define the metabolic niche of microorganisms. Such genes are generally considered to have undergone high rates of lateral gene transfer. Many gene sequences formed clades that were identical, or similar, to the tree constructed using large numbers of genes from the stable core of the genome. Surprisingly, such lateral transfer events were readily identified and quantifiable, occurring only a relatively small number of times in the archaeal domain of life. By placing gene acquisition events into a temporal framework, the rates by which new metabolic genes were acquired can be quantified. The highest lateral transfer rates were among cytochrome oxidase genes that use oxygen as a terminal electron acceptor (with a total of 12-14 lateral transfer events, or 3.4-4.0 events per billion years, across the entire archaeal domain). Genes involved in sulfur or nitrogen metabolism had much lower rates, on the order of one lateral transfer event per billion years. This suggests that lateral transfer rates of key metabolic proteins are rare and not rampant.

An expansion of age constraints for microbial clades that lack a conventional fossil record using phylogenomic dating.

Most microbial taxa lack a conventional microfossil or biomarker record, and so we currently have little information regarding how old most microbial clades and their associated traits are. Building on the previously published oxygen age constraint, two new age constraints are proposed based on the ability of microbial clades to metabolize chitin and aromatic compounds derived from lignin. Using the archaeal domain of life as a test case, phylogenetic analyses, along with published metabolic and genetic data, showed that members of the Halobacteriales and Thermococcales are able to metabolize chitin. Ancestral state reconstruction combined with phylogenetic analysis of the genes underlying chitin degradation predicted that the ancestors of these two groups were also likely able to metabolize chitin or chitin-related compounds. These two clades were therefore assigned a maximum age of 1.0 Ga (when chitin likely first appeared). Similar analyses also predicted that the ancestor to the Sulfolobus solfataricus-Sulfolobus islandicus clade was able to metabolize phenol using catechol dioxygenase, so this clade was assigned a maximum age of 475 Ma. Inferred ages of archaeal clades using relaxed molecular clocks with the new age constraints were consistent with those inferred with the oxygen age constraints. This work expands our current toolkit to include Paleoproterozoic, Neoproterozoic, and Paleozoic age constraints, and should aid in our ability to phylogenetically reconstruct the antiquity of a wide array of microbial clades and their associated morphological and biogeochemical traits, spanning deep geologic time. Such hypotheses-although built upon evolutionary inferences-are fundamentally testable.

Not so old Archaea - the antiquity of biogeochemical processes in the archaeal domain of life.

Since the archaeal domain of life was first recognized, it has often been assumed that Archaea are ancient, and harbor primitive traits. In fact, the names of the major archaeal lineages reflect our assumptions regarding the antiquity of their traits. Ancestral state reconstruction and relaxed molecular clock analyses using newly articulated oxygen age constraints show that although the archaeal domain itself is old, tracing back to the Archean eon, many clades and traits within the domain are not ancient or primitive. Indeed many clades and traits, particularly in the Euryarchaeota, were inferred to be Neoproterozoic or Phanerozoic in age. Both Eury- and Crenarchaeota show increasing metabolic and physiological diversity through time. Early archaeal microbial communities were likely limited to sulfur reduction and hydrogenotrophic methanogenesis, and were confined to high-temperature geothermal environments. However, after the appearance of atmospheric oxygen, nodes containing a wide variety of traits (sulfate and thiosulfate reduction, sulfur oxidation, sulfide oxidation, aerobic respiration, nitrate reduction, mesophilic methanogenesis in sedimentary environments) appear, first in environments containing terrestrial Crenarchaeota in the Meso/Neoproterozoic followed by environments containing marine Euryarchaeota in the Neoproterozoic and Phanerozoic. This provides phylogenetic evidence for increasing complexity in the biogeochemical cycling of C, N, and S through geologic time, likely as a consequence of microbial evolution and the gradual oxygenation of various compartments within the biosphere. This work has implications not only for the large-scale evolution of microbial communities and biogeochemical processes, but also for the interpretation of microbial biosignatures in the ancient rock record.

Phylogenomic dating--a method of constraining the age of microbial taxa that lack a conventional fossil record.

A phylogenomic dating approach was used to identify potential age constraints for multiple archaeal groups, many of which have no fossil, isotopic, or biomarker record. First, well-resolved phylogenetic trees were inferred with the use of multiple gene sequences obtained from whole genome sequences. Next, the ability to use oxygen as a terminal electron acceptor was coded into characters, and ancestral state reconstruction was used to identify clades with taxa that metabolize oxygen and likely had an aerobic ancestor. Next, the habitat of the ancestor was inferred. If the local presence of Cyanobacteria could be excluded from the putative ancestral habitat, then these clades would have originated after the rise in atmospheric oxygen 2.32 Ga. With this method, an upper age of 2.32 Ga (an "oxygen age constraint") is proposed for four major archaeal clades: the Sulfolobales, Thermoplasmatales, Thermoproteus neutrophilus/Pyrobaculum spp., and the Thermoproteales. It was also shown that the halophilic archaea likely had an aerobic common ancestor, yet the possibility of local oxygen oases before oxygenation of the atmosphere could not be formally rejected. Thus, an oxygen age constraint was not assessed for this group. This work suggests that many archaeal groups are not as ancient as many in the research community have previously assumed, and it provides a new method for establishing upper age constraints for major microbial groups that lack a conventional fossil record.

Phylogenomic dating--the relative antiquity of archaeal metabolic and physiological traits.

Ancestral trait reconstruction was used to identify the relative ancestry of metabolic and physiological traits in the archaeal domain of life. First, well-resolved phylogenetic trees were inferred with multiple gene sequences obtained from whole genome sequences. Next, metabolic and physiological traits were coded into characters, and ancestral state reconstruction was used to identify ancient and derived traits. Traits inferred to be ancient included sulfur reduction, methanogenesis, and hydrogen oxidation. By using the articulation of the "oxygen age constraint," several other traits were inferred to have arisen at or after 2.32 Ga: aerobic respiration, nitrate reduction, sulfate reduction, thiosulfate reduction, sulfur oxidation, and sulfide oxidation. Complex organic metabolism appeared to be nearly as ancient as autotrophy. Hyperthermophily was ancestral, while hyperacidophily and extreme halophily likely arose after 2.32 Ga. The ancestral euryarchaeote was inferred to have been a hyperthermophilic marine methanogen that lived in a deep-sea hydrothermal vent. In contrast, the ancestral crenarchaeote was most likely a hyperthermophilic sulfur reducer that lived in a slightly acidic terrestrial environment, perhaps a fumarole. Cross-colonization of these habitats may not have occurred until after 2.32 Ga, which suggests that both archaeal lineages exhibited niche specialization on early Earth for a protracted period of time.

Microbial composition of near-boiling silica-depositing thermal springs throughout Yellowstone National Park.

The extent of hyperthermophilic microbial diversity associated with siliceous sinter (geyserite) was characterized in seven near-boiling silica-depositing springs throughout Yellowstone National Park using environmental PCR amplification of small-subunit rRNA genes (SSU rDNA), large-subunit rDNA, and the internal transcribed spacer (ITS). We found that Thermocrinis ruber, a member of the order Aquificales, is ubiquitous, an indication that primary production in these springs is driven by hydrogen oxidation. Several other lineages with no known close relatives were identified that branch among the hyperthermophilic bacteria. Although they all branch deep in the bacterial tree, the precise phylogenetic placement of many of these lineages is unresolved at this time. While some springs contained a fair amount of phylogenetic diversity, others did not. Within the same spring, communities in the subaqueous environment were not appreciably different than those in the splash zone at the edge of the pool, although a greater number of phylotypes was found along the pool's edge. Also, microbial community composition appeared to have little correlation with the type of sinter morphology. The number of cell morphotypes identified by fluorescence in situ hybridization and scanning electron microscopy was greater than the number of phylotypes in SSU clone libraries. Despite little variation in Thermocrinis ruber SSU sequences, abundant variation was found in the hypervariable ITS region. The distribution of ITS sequence types appeared to be correlated with distinct morphotypes of Thermocrinis ruber in different pools. Therefore, species- or subspecies-level divergences are present but not detectable in highly conserved SSU sequences.