An overview of topic modeling and its current applications in bioinformatics

Abstract

BackgroundWith the rapid accumulation of biological datasets, machine learning methods designed to automate data analysis are urgently needed. In recent years, so-called topic models that originated from the field of natural language processing have been receiving much attention in bioinformatics because of their interpretability. Our aim was to review the application and development of topic models for bioinformatics.DescriptionThis paper starts with the description of a topic model, with a focus on the understanding of topic modeling. A general outline is provided on how to build an application in a topic model and how to develop a topic model. Meanwhile, the literature on application of topic models to biological data was searched and analyzed in depth. According to the types of models and the analogy between the concept of document-topic-word and a biological object (as well as the tasks of a topic model), we categorized the related studies and provided an outlook on the use of topic models for the development of bioinformatics applications.ConclusionTopic modeling is a useful method (in contrast to the traditional means of data reduction in bioinformatics) and enhances researchers’ ability to interpret biological information. Nevertheless, due to the lack of topic models optimized for specific biological data, the studies on topic modeling in biological data still have a long and challenging road ahead. We believe that topic models are a promising method for various applications in bioinformatics research.

The Personality Lexicon in Brazilian Portuguese: Studies with Natural Language

“…The LDA was introduced by Blei et al (2003) and became a popular model for uncovering latent topics in large text corpora and other kinds of discrete data (Griffiths, Steyvers, & Tenenbaum, 2007;Grün & Hornik, 2011;Kosinski et al, 2016;Liu, Tang, Dong, Yao, & Zhou, 2016;Poldrack et al, 2012). According to Blei and Lafferty (n.d.):…”

Topic Modeling: A Comprehensive Review

“…This helps science foundation administrators to better understand the content and context of funding portfolios in order to help promote future science funding plans. A topic model is presented for simultaneously modeling papers, authors, and publication venues known as Author conference topic model [73]. It includes data of 14134 authors 10716 papers and 1434 conferences from Arnet Miner.…”

Capturing cell type-specific chromatin structural patterns by applying topic modeling to single-cell Hi-C data

“…With this generative model, the LDA identifies a set of topics that have most likely generated a given collection of documents. In the end, LDA generates two matrices, one that describes the relationship between topics and document and the other describing the relationship between topics and words [9]. In the first matrix, topics with the highest contributions are those that best characterize a document.…”