Fast, Consistent Tokenization of Natural Language Text

Mullen, Lincoln; Benoit, Kenneth; Keyes, Os; Selivanov, Dmitry; Arnold, Jeffrey

doi:10.21105/joss.00655

Cited by 91 publications

(54 citation statements)

References 8 publications

Supporting

Mentioning

Contrasting

Unclassified

Order By: Relevance

“…All CDR3s-amino acid (a.a.) from each patient's TCR repertoire were deconstructed into overlapping subsequences ("k-mers") of length 3 (k = 3) using the tokenizers R package (29). Then, all k-mers were condensed into a k-mer frequency distribution containing each k-mer's frequency across all CDR3s of a given repertoire.…”

Section: K-mer-based Cdr3 Subsequence Analysismentioning

confidence: 99%

The TCR Repertoire Reconstitution in Multiple Sclerosis: Comparing One-Shot and Continuous Immunosuppressive Therapies

et al. 2020

View full text Add to dashboard Cite

has not yet been established hindering insight into the posttreatment TCR landscape of MS patients. To address this important knowledge gap, we tracked peripheral T-cell subpopulations (naive and memory CD4+ and CD8+) across 15 RRMS patients before and after 2 years of continuous treatment (NTZ) and a single treatment course (AHSCT) by high-throughput TCRβ sequencing. We found that the two MS treatments left treatment-specific multidimensional traces in patient TCRβ repertoire dynamics with respect to clonal expansion, clonal diversity, and repertoire architecture. Comparing MS TCR sequences with published datasets suggested that the majority of public TCRs belonged to virus-associated sequences. In summary, applying multidimensional computational immunology to a TCRβ dataset of treated MS patients, we show that qualitative changes of TCRβ repertoires encode treatment-specific information that may be relevant for future clinical trials monitoring and personalized MS follow-up, diagnosis, and treatment regimens.

show abstract

Section: K-mer-based Cdr3 Subsequence Analysismentioning

confidence: 99%

The TCR Repertoire Reconstitution in Multiple Sclerosis: Comparing One-Shot and Continuous Immunosuppressive Therapies

et al. 2020

View full text Add to dashboard Cite

show abstract

“…Consecutively, we analyzed the data using text mining methods (Silge and Robinson, 2016 ) to estimate the frequency and the association of body patterns, such as radar charts and correlations with Bonferroni correction for multiple comparisons. All the analyses were conducted in (RStudio Team, 2015 ) v1.2.1335® (RStudio Team, 2015 ) using the packages FactoMineR v1.42 (Lê et al, 2008 ), tidyverse v1.2.1 (Wickham et al, 2019a ), tidytext v0.2.2 (Silge and Robinson, 2016 ), dplyr v0.8.3 (Wickham et al, 2019b ), widyr v0.1.2 (Robinson, 2019 ), tokenizers v0.2.1 (Mullen et al, 2018 ), quanteda v1.5.1 (Benoit et al, 2018 ) and igraph v1.2.4.1 (Csardi and Nepusz, 2006 ).…”

Section: Methodsmentioning

confidence: 99%

Dynamic Courtship Signals and Mate Preferences in Sepia plangon

2020

View full text Add to dashboard Cite

Communication in cuttlefish includes rapid changes in skin coloration and texture, body posture and movements, and potentially polarized signals. The dynamic displays are fundamental for mate choice and agonistic behavior. We analyzed the reproductive behavior of the mourning cuttlefish Sepia plangon in the laboratory. Mate preference was analyzed via choice assays ( n = 33) under three sex ratios, 1 male (M): 1 female (F), 2M:1F, and 1M:2F. We evaluated the effect of modifying polarized light from the arms stripes and ambient light with polarized and unpolarized barriers between the cuttlefish. Additionally, to assess whether a particular trait was a determinant for mating, we used 3D printed cuttlefish dummies. The dummies had different sets of visual signals: two sizes (60 or 90 mm mantle length), raised or dropped arms, high or low contrast body coloration, and polarized or unpolarized filters to simulate the arms stripes. Frequency and duration (s) of courtship displays, mating, and agonistic behaviors were analyzed with GLM and ANOVAs. The behaviors, body patterns, and their components were integrated into an ethogram to describe the reproductive behavior of S. plangon . We identified 18 body patterns, 57 body patterns components, and three reproductive behaviors (mating, courtship, and mate guarding). Only sex ratio had a significant effect on courtship frequency, and the male courtship success rate was 80%. Five small (ML < 80 mm) males showed the dual-lateral display to access mates while avoiding fights with large males; this behavior is characteristic of male “sneaker” cuttlefish. Winner males showed up to 17 body patterns and 33 components, whereas loser males only showed 12 patterns and 24 components. We identified 32 combinations of body patterns and components that tended to occur in a specific order and were relevant for mating success in males. Cuttlefish were visually aware of the 3D-printed dummies; however, they did not start mating or agonistic behavior toward the dummies. Our findings suggest that in S. plangon , the dynamic courtship displays with specific sequences of visual signals, and the sex ratio are critical for mate choice and mating success.

show abstract

“…Using the combination of rentrez [8], easyPubMed [9], pubmed.mineR [10], tokenizers [11] and blogdown [12] R packages, and Hugo [13] it…”

Section: Methodsmentioning

confidence: 99%

PlatCOVID: A Novel Web Tool to Analyze, Curate and Share COVID-19 Literature

Brandão

Agrelli

Bernardo

et al. 2020

Preprint

View full text Add to dashboard Cite

BackgroundIn the attempt to face the COVID-19 pandemic, the global scientific community has been expending great efforts to produce useful and reliable data aiming to help patients, physicians and guiding public health policies. A huge amount of information is being released every week, making impossible for a single person (or even for a research group) to read everything and get constantly updated on the scientific literature concerning COVID-19 and its etiological agent, SARS-CoV-2. Therefore, we developed PlatCOVID (www.platcovid.com), a Web platform designed to analyze, cluster, classify and discuss COVID-19 literature available on LitCovid (NCBI). ResultsPlatCOVID has been created as a novel COVID-19 hub able to add features of text mining and syntax analyses methods, such as word and sentence atomization and tokenization, clusterization and classification. The main division of the literature comprehends five categories: 1) Diagnosis; 2) Epidemiology; 3) Clinical, Signs & Symptoms; 4) Transmission; and 5) Treatment & Prevention. Consequently, it is possible to reduce the amount of text to be read with minimal loss of information, identifying target subjects by mining as new insights arise, enhancing data analysis efficiency. PlatCOVID has been designed with central panels (Gene, Drug and Tissue panels) to easily gather and share with the scientific community important COVID-19 information.ConclusionsAlthough most of the text mining and syntax analysis is made by using automated computing processes, the final results must to be humanly curated. With this in mind, PlatCOVID allows researchers to be part of our effort to curate, analyze, discuss and rate each matter of interest (helpus.platcovid.com). We welcome user feedback for further enhancement.

show abstract

Fast, Consistent Tokenization of Natural Language Text

Cited by 91 publications

References 8 publications

The TCR Repertoire Reconstitution in Multiple Sclerosis: Comparing One-Shot and Continuous Immunosuppressive Therapies

The TCR Repertoire Reconstitution in Multiple Sclerosis: Comparing One-Shot and Continuous Immunosuppressive Therapies

Dynamic Courtship Signals and Mate Preferences in Sepia plangon

PlatCOVID: A Novel Web Tool to Analyze, Curate and Share COVID-19 Literature

Contact Info

Product

Resources

About