Your browser doesn't support javascript.
loading
Applications and Comparison of Dimensionality Reduction Methods for Microbiome Data.
Armstrong, George; Rahman, Gibraan; Martino, Cameron; McDonald, Daniel; Gonzalez, Antonio; Mishne, Gal; Knight, Rob.
Afiliação
  • Armstrong G; Department of Pediatrics, School of Medicine, University of California, San Diego, La Jolla, CA, United States.
  • Rahman G; Bioinformatics and Systems Biology Program, University of California, San Diego, La Jolla, CA, United States.
  • Martino C; Department of Pediatrics, School of Medicine, University of California, San Diego, La Jolla, CA, United States.
  • McDonald D; Bioinformatics and Systems Biology Program, University of California, San Diego, La Jolla, CA, United States.
  • Gonzalez A; Department of Pediatrics, School of Medicine, University of California, San Diego, La Jolla, CA, United States.
  • Mishne G; Bioinformatics and Systems Biology Program, University of California, San Diego, La Jolla, CA, United States.
  • Knight R; Center for Microbiome Innovation, Jacobs School of Engineering, University of California, San Diego, La Jolla, CA, United States.
Front Bioinform ; 2: 821861, 2022.
Article em En | MEDLINE | ID: mdl-36304280
Dimensionality reduction techniques are a key component of most microbiome studies, providing both the ability to tractably visualize complex microbiome datasets and the starting point for additional, more formal, statistical analyses. In this review, we discuss the motivation for applying dimensionality reduction techniques, the special characteristics of microbiome data such as sparsity and compositionality that make this difficult, the different categories of strategies that are available for dimensionality reduction, and examples from the literature of how they have been successfully applied (together with pitfalls to avoid). We conclude by describing the need for further development in the field, in particular combining the power of phylogenetic analysis with the ability to handle sparsity, compositionality, and non-normality, as well as discussing current techniques that should be applied more widely in future analyses.
Palavras-chave

Texto completo: 1 Base de dados: MEDLINE Idioma: En Revista: Front Bioinform Ano de publicação: 2022 Tipo de documento: Article País de afiliação: Estados Unidos

Texto completo: 1 Base de dados: MEDLINE Idioma: En Revista: Front Bioinform Ano de publicação: 2022 Tipo de documento: Article País de afiliação: Estados Unidos