No abstract

BackgroundTopic modelling is an active research field in machine learning. While mainly used to build models from unstructured textual data, it offers an effective means of data mining where samples represent documents, and different biological endpoints or omics data represent words. Latent Dirichlet Allocation (LDA) is the most commonly used topic modelling method across a wide number of technical fields. However, model development can be arduous and tedious, and requires burdensome and systematic sensitivity studies in order to find the best set of model parameters. Often, time-consuming subjective evaluations are needed to compare models. Currently, research has yielded no easy way to choose the proper number of topics in a model beyond a major iterative approach.Methods and resultsBased on analysis of variation of statistical perplexity during topic modelling, a heuristic approach is proposed in this study to estimate the most appropriate number of topics. Specifically, the rate of perplexity change (RPC) as a function of numbers of topics is proposed as a suitable selector. We test the stability and effectiveness of the proposed method for three markedly different types of grounded-truth datasets: Salmonella next generation sequencing, pharmacological side effects, and textual abstracts on computational biology and bioinformatics (TCBB) from PubMed.ConclusionThe proposed RPC-based method is demonstrated to choose the best number of topics in three numerical experiments of widely different data types, and for databases of very different sizes. The work required was markedly less arduous than if full systematic sensitivity studies had been carried out with number of topics as a parameter. We understand that additional investigation is needed to substantiate the method's theoretical basis, and to establish its generalizability in terms of dataset characteristics.

IMPORTANCETo our knowledge, a set of well-defined diagnostic criteria is not yet developed for the diagnosis of Vogt-Koyanagi-Harada (VKH) disease.OBJECTIVE To develop and evaluate a set of diagnostic criteria for VKH disease using data from Chinese patients. DESIGN, SETTING, AND PARTICIPANTS This case-control study reviewed medical records of patients from a tertiary referral center between October 2011 and October 2016. Data from 634 patients with VKH disease and 623 patients with non-VKH uveitis from southern China were used to develop the Diagnostic Criteria for VKH Disease (DCV). Data from an additional group of 537 patients with a definite VKH disease diagnosis and 525 patients with non-VKH uveitis from northern China were used to evaluate the diagnostic criteria. MAIN OUTCOMES AND MEASURES Sensitivity, specificity, positive predictive value (PPV), negative predictive value (NPV), and receiver operating characteristic. RESULTSOf the 1257 patients used to construct the DCV, 665 (52.9%) were male, and the mean (SD) age at disease onset was 38.6 (13.6) years. The 3-class model and 21 clinical findings were selected by latent class analysis. Variables with a high positive rate in the early-phase or late-phase VKH group or high specificity constituted essential parameters. Constellations of these essential parameters constructed the DCV. The sensitivity and NPV of the DCV were higher than those of the Revised Diagnostic Criteria for VKH Disease (RDC)

BackgroundBoth adolescent substance use and adolescent depression are major public health problems, and have the tendency to co-occur. Thousands of articles on adolescent substance use or depression have been published. It is labor intensive and time consuming to extract huge amounts of information from the cumulated collections. Topic modeling offers a computational tool to find relevant topics by capturing meaningful structure among collections of documents.MethodsIn this study, a total of 17,723 abstracts from PubMed published from 2000 to 2014 on adolescent substance use and depression were downloaded as objects, and Latent Dirichlet allocation (LDA) was applied to perform text mining on the dataset. Word clouds were used to visually display the content of topics and demonstrate the distribution of vocabularies over each topic.ResultsThe LDA topics recaptured the search keywords in PubMed, and further discovered relevant issues, such as intervention program, association links between adolescent substance use and adolescent depression, such as sexual experience and violence, and risk factors of adolescent substance use, such as family factors and peer networks. Using trend analysis to explore the dynamics of proportion of topics, we found that brain research was assessed as a hot issue by the coefficient of the trend test.ConclusionsTopic modeling has the ability to segregate a large collection of articles into distinct themes, and it could be used as a tool to understand the literature, not only by recapturing known facts but also by discovering other relevant topics.Electronic supplementary materialThe online version of this article (doi:10.1186/s12889-016-2932-1) contains supplementary material, which is available to authorized users.

BackgroundThe big data moniker is nowhere better deserved than to describe the ever-increasing prodigiousness and complexity of biological and medical datasets. New methods are needed to generate and test hypotheses, foster biological interpretation, and build validated predictors. Although multivariate techniques such as cluster analysis may allow researchers to identify groups, or clusters, of related variables, the accuracies and effectiveness of traditional clustering methods diminish for large and hyper dimensional datasets. Topic modeling is an active research field in machine learning and has been mainly used as an analytical tool to structure large textual corpora for data mining. Its ability to reduce high dimensionality to a small number of latent variables makes it suitable as a means for clustering or overcoming clustering difficulties in large biological and medical datasets.ResultsIn this study, three topic model-derived clustering methods, highest probable topic assignment, feature selection and feature extraction, are proposed and tested on the cluster analysis of three large datasets: Salmonella pulsed-field gel electrophoresis (PFGE) dataset, lung cancer dataset, and breast cancer dataset, which represent various types of large biological or medical datasets. All three various methods are shown to improve the efficacy/effectiveness of clustering results on the three datasets in comparison to traditional methods. A preferable cluster analysis method emerged for each of the three datasets on the basis of replicating known biological truths.ConclusionTopic modeling could be advantageously applied to the large datasets of biological or medical research. The three proposed topic model-derived clustering methods, highest probable topic assignment, feature selection and feature extraction, yield clustering improvements for the three different data types. Clusters more efficaciously represent truthful groupings and subgroupings in the data than traditional methods, suggesting that topic model-based methods could provide an analytic advancement in the analysis of large biological or medical datasets.

Species identification via DNA barcodes is contributing greatly to current bioinventory efforts. The initial, and widely accepted, proposal was to use the protein-coding cytochrome c oxidase subunit I (COI) region as the standard barcode for animals, but recently non-coding internal transcribed spacer (ITS) genes have been proposed as candidate barcodes for both animals and plants. However, achieving a robust alignment for non-coding regions can be problematic. Here we propose two new methods (DV-RBF and FJ-RBF) to address this issue for species assignment by both coding and non-coding sequences that take advantage of the power of machine learning and bioinformatics. We demonstrate the value of the new methods with four empirical datasets, two representing typical protein-coding COI barcode datasets (neotropical bats and marine fish) and two representing non-coding ITS barcodes (rust fungi and brown algae). Using two random sub-sampling approaches, we demonstrate that the new methods significantly outperformed existing Neighbor-joining (NJ) and Maximum likelihood (ML) methods for both coding and non-coding barcodes when there was complete species coverage in the reference dataset. The new methods also out-performed NJ and ML methods for non-coding sequences in circumstances of potentially incomplete species coverage, although then the NJ and ML methods performed slightly better than the new methods for protein-coding barcodes. A 100% success rate of species identification was achieved with the two new methods for 4,122 bat queries and 5,134 fish queries using COI barcodes, with 95% confidence intervals (CI) of 99.75–100%. The new methods also obtained a 96.29% success rate (95%CI: 91.62–98.40%) for 484 rust fungi queries and a 98.50% success rate (95%CI: 96.60–99.37%) for 1094 brown algae queries, both using ITS barcodes.

No abstract

No abstract