In microbiome/miaTime: Microbiome Time Series Analysis

#| label: setup
#| include: false

library(knitr)
knitr::opts_chunk$set(
    collapse = TRUE,
    comment = "#>",
    cache = TRUE
)

Introduction

miaTime is a package in the r BiocStyle::Biocpkg("mia") family, providing tools for time series manipulation using the r BiocStyle::Biocpkg("TreeSummarizedExperiment") data container.

Installation

miaTime is hosted on Bioconductor, and can be installed using via BiocManager.

#| label: install
#| eval: false

BiocManager::install("miaTime")

Load the package

Once installed, miaTime is made available in the usual way.

#| label: load_package

library(miaTime)

Divergence between time points

miaTime offers functions to calculate divergences. These can be calculated based on samples and their corresponding base time point, e.g., first sample of time series. Moreover, divergences can be calculated in rolling basis meaning that a sample is compared to previous ith sample.

Divergences can be calculated with get*Divergence() functions. In the example below, for each subject, we calculate the divergence of their samples by comparing them to the first time point.

#| label: base_divergence

data(hitchip1006)
tse <- hitchip1006

res <- getBaselineDivergence(
    tse, time.col = "time", group = "sample", name = "baseline")
res |> head()

A more convenient and preferred approach is to store the values directly in colData using the get*Divergence() functions. In the example below, we calculate stepwise divergences with a lag of 1, meaning that for each sample, the divergence is calculated by comparing it to the previous time point for the same subject.

#| label: time_divergence

tse <- addStepwiseDivergence(tse, time.col = "time")
colData(tse)

Visualize time series

We can visualize time series data with r BiocStyle::Biocpkg("miaViz"). Below, we visualize 2 most abundant taxa.

#| label: plot_series

library(miaViz)

p <- plotSeries(tse, x = "time", y = getTop(tse, 5))
p

See articles for more detailed example workflows.

Session info

#| label: session_info

sessionInfo()

microbiome/miaTime documentation built on Nov. 4, 2024, 10:23 p.m.