In jaspershen/massstat: Statistical analysis for LC-MS-based metabolomics data

knitr::opts_chunk$set(
  collapse = TRUE,
  comment = "#>",
  warning = FALSE,
  message = TRUE,
  out.width = "100%"
)

Calculate correlation

library(massstat)
library(massdataset)
library(magrittr)
library(dplyr)

data("liver_aging_pos")
liver_aging_pos

qc_id <-
  liver_aging_pos %>%
  activate_mass_dataset(what = "sample_info") %>%
  dplyr::filter(group == "QC") %>%
  dplyr::pull(sample_id)

object <-
  mutate_rsd(liver_aging_pos, according_to_samples = qc_id)

###only remain the features with rt > 100, mz > 150 and rsd < 30
object <-
  object %>%
  activate_mass_dataset(what = "variable_info") %>%
  dplyr::filter(rt > 100) %>%
  dplyr::filter(mz > 150) %>%
  dplyr::filter(rsd < 30)

##only remain the week 24 samples
object <-
  object %>%
  activate_mass_dataset(what = "sample_info") %>%
  dplyr::filter(group == "24W")

dim(object)

object <-
  object %>%
  `+`(1) %>%
  log(10) %>%
  scale_data(method = "auto")

cor_data <-
  object %>%
  cor_mass_dataset(margin = "variable", data_type = "wider")

head(cor_data$correlation[,1:5])
head(cor_data$p_value[,1:5])
head(cor_data$n[,1:5])

cor_data <-
  object %>%
  cor_mass_dataset(margin = "variable", data_type = "longer")

head(cor_data)

Calculate distance

library(massstat)
library(massdataset)
library(tidyverse)
data("expression_data")
data("sample_info")
data("sample_info_note")
data("variable_info")
data("variable_info_note")
object =
  create_mass_dataset(
    expression_data = expression_data,
    sample_info = sample_info,
    variable_info = variable_info,
    sample_info_note = sample_info_note,
    variable_info_note = variable_info_note
  )
object
x =
  object %>%
  log(2) %>%
  scale()
variable_distance <-
  dist_mass_dataset(x = x, margin = "variable")
head(as.matrix(variable_distance)[, 1:5])
sample_distance <-
  dist_mass_dataset(x = x, margin = "sample")
head(as.matrix(sample_distance)[, 1:5])