![]() ![]() ![]() The mSet class and its child classes must include four data slots: conc_table, sample_table, feature_data, and experiment_data. However, a method defined in any of the child classes cannot be applied to its siblings. All mSet methods can be applied to each of its child classes. ![]() The mSet is a template for other classes so it is not directly callable. All the classes above inherits from a virtual class called mSet. The MultxSet can be used as a container for general experiment data such as dietary data, clinical values, anthropometric data, or ELISA data. Metabase currently has four main data container classes designed for experiments with specific purposes, MetabolomicsSet, LipidomicsSet, ProteomicsSet, MicrobiomeSet, and a generic class MultxSet. The Metabase package takes the data structure design of the two package, with an emphasis on the data analysis, manipulation, hypothesis testing and visualization. The former is the infrastructural package of bioconductor while the latter is a package designated for microbiome data analysis. The Metabase package is inspired by the bioconductor package Biobase and phyloseq. In a microbiome study, the feature_table contains the taxonomy labels in each level from kingdom to species. In a proteomics study, it may contain the protein name, sequence, gene name, database ID such as emsembl ID. In a metabolomics/lipidomics study, it may contain the molecule weight, m/z value, ion mode, annotation(metabolite name), and chemical CID such as PubChem CID. While the feature_data contains the additional information of each feature. The sample_table usually contains the study design information, such as treatment group, dosage, age, and gender. It can be referred to metabolites in a metabolomics study, lipids in a lipidomcis study, proteins/peptides in a proteomics study, and OTUs, genera, or phylums in a microbiome study. In this package and in the rest of this vignette, the three parts of the dataset are referred as conc_table, sample_table, and feature_data.įeature is a generalized term. A numeric \(m \times n\) matrix that each number represents the concentration/abundance of the \(j^\) sample, a \(n \times k\) table of the sample information, and a \(m \times l\) table of the feature information. A study with n samples and m features usually contains three portions of the data. High-throughput ‘omics’ studies such as metabolomics, lipidomics, protomics and microbiome usually share a very similar data structure when they are applied in a specific population to answer specific questions. 1 Data structure of high through-put experiments ![]()
0 Comments
Leave a Reply. |
Details
AuthorWrite something about yourself. No need to be fancy, just an overview. ArchivesCategories |