Dataset Information

Extraction of phenotypic traits from taxonomic descriptions for the tree of life using natural language processing.

ABSTRACT:

Premise of the study

Phenotypic data sets are necessary to elucidate the genealogy of life, but assembling phenotypic data for taxa across the tree of life can be technically challenging and prohibitively time consuming. We describe a semi-automated protocol to facilitate and expedite the assembly of phenotypic character matrices of plants from formal taxonomic descriptions. This pipeline uses new natural language processing (NLP) techniques and a glossary of over 9000 botanical terms.

Methods and results

Our protocol includes the Explorer of Taxon Concepts (ETC), an online application that assembles taxon-by-character matrices from taxonomic descriptions, and MatrixConverter, a Java application that enables users to evaluate and discretize the characters extracted by ETC. We demonstrate this protocol using descriptions from Araucariaceae.

Conclusions

The NLP pipeline unlocks the phenotypic data found in taxonomic descriptions and makes them usable for evolutionary analyses.

SUBMITTER: Endara L

PROVIDER: S-EPMC5895189 | biostudies-literature | 2018 Mar

REPOSITORIES: biostudies-literature

ACCESS DATA

Publications

Extraction of phenotypic traits from taxonomic descriptions for the tree of life using natural language processing.

Endara Lorena L Cui Hong H Burleigh J Gordon JG

Applications in plant sciences 20180331 3

<h4>Premise of the study</h4>Phenotypic data sets are necessary to elucidate the genealogy of life, but assembling phenotypic data for taxa across the tree of life can be technically challenging and prohibitively time consuming. We describe a semi-automated protocol to facilitate and expedite the assembly of phenotypic character matrices of plants from formal taxonomic descriptions. This pipeline uses new natural language processing (NLP) techniques and a glossary of over 9000 botanical terms.<h ...[more]

PMID: 29732265

Similar Datasets

Project description:Widely used microbial taxonomies, such as the NCBI taxonomy, are based on a combination of sequence homology among conserved genes and historically accepted taxonomies, which were developed based on observable traits such as morphology and physiology. A recently proposed alternative taxonomy database, the Genome Taxonomy Database (GTDB), incorporates only sequence homology of conserved genes and attempts to partition taxonomic ranks such that each rank implies the same amount of evolutionary distance, regardless of its position on the phylogenetic tree. This provides the first opportunity to completely separate taxonomy from traits and therefore to quantify how taxonomic rank corresponds to traits across the microbial tree of life. We quantified the relative abundances of clusters of orthologous group functional categories (COG-FCs) as a proxy for traits within the lineages of 13,735 cultured and uncultured microbial lineages from a custom-curated genome database. On average, 41.4% of the variation in COG-FC relative abundance is explained by taxonomic rank, with domain, phylum, class, order, family, and genus explaining, on average, 3.2%, 14.6%, 4.1%, 9.2%, 4.8%, and 5.5% of the variance, respectively (P < 0.001 for all). To our knowledge, this is the first work to quantify the variance in metabolic potential contributed by individual taxonomic ranks. A qualitative comparison between the COG-FC relative abundances and genus-level phylogenies, generated from published concatenated protein sequence alignments, further supports the idea that metabolic potential is taxonomically coherent at higher taxonomic ranks. The quantitative analyses presented here characterize the integral relationship between diversification of microbial lineages and the metabolisms which they host.IMPORTANCE Recently, there has been great progress in defining a complete taxonomy of bacteria and archaea, which has been enabled by improvements in DNA sequencing technology and new bioinformatic techniques. A new, algorithmically defined microbial tree of life describes those linkages, relying solely on genetic data, which raises the issue of how microbial traits relate to taxonomy. Here, we adopted cluster of orthologous group functional categories as a scheme to describe the genomic contents of microbes, a method that can be applied to any microbial lineage for which genomes are available. This simple approach allows quantitative comparisons between microbial genomes with different gene compositions from across the microbial tree of life. Our observations demonstrate statistically significant patterns in cluster of orthologous group functional categories at taxonomic levels that span the range from domain to genus.

Project description:BackgroundNatural language processing (NLP) offers significantly faster variable extraction compared to traditional human extraction but cannot interpret complicated notes as well as humans can. Thus, we hypothesized that an "NLP-assisted" extraction system, which uses humans for complicated notes and NLP for uncomplicated notes, could produce faster extraction without compromising accuracy.ObjectiveThe aim of this study was to develop and pilot an NLP-assisted extraction system to leverage the strengths of both human and NLP extraction of prostate cancer Gleason scores.MethodsWe collected all available clinical and pathology notes for prostate cancer patients in an unselected academic biobank cohort. We developed an NLP system to extract prostate cancer Gleason scores from both clinical and pathology notes. Next, we designed and implemented the NLP-assisted extraction system algorithm to categorize notes into "uncomplicated" and "complicated" notes. Uncomplicated notes were assigned to NLP extraction and complicated notes were assigned to human extraction. We randomly reviewed 200 patients to assess the accuracy and speed of our NLP-assisted extraction system and compared it to NLP extraction alone and human extraction alone.ResultsOf the 2051 patients in our cohort, the NLP system extracted a prostate surgery Gleason score from 1147 (55.92%) patients and a prostate biopsy Gleason score from 1624 (79.18%) patients. Our NLP-assisted extraction system had an overall accuracy rate of 98.7%, which was similar to the accuracy of human extraction alone (97.5%; P=.17) and significantly higher than the accuracy of NLP extraction alone (95.3%; P<.001). Moreover, our NLP-assisted extraction system reduced the workload of human extractors by approximately 95%, resulting in an average extraction time of 12.7 seconds per patient (vs 256.1 seconds per patient for human extraction alone).ConclusionsWe demonstrated that an NLP-assisted extraction system was able to achieve much faster Gleason score extraction compared to traditional human extraction without sacrificing accuracy.

Dataset Information

Extraction of phenotypic traits from taxonomic descriptions for the tree of life using natural language processing.

Premise of the study

Methods and results

Conclusions

Publications

Extraction of phenotypic traits from taxonomic descriptions for the tree of life using natural language processing.

Similar Datasets

OmicsDI is part of the ELIXIR infrastructure

Tweets