MPC non-targeted metabolomics normalization workflow

This repository contains all files necessary to perform different normalization steps for non-targeted data generated in the MPC at HMGU. Files shall be copied to the respective project and executed within the project folder. In order to insure correct functionality, please be advised to install the neccesary packages as follows:

R Libraries

ipak <- function(pkg){

new.pkg <- pkg[!(pkg %in% installed.packages()[, "Package"])]

if (length(new.pkg)) install.packages(new.pkg, dependencies = TRUE)

sapply(pkg, require, character.only = TRUE)

};

packages <- c("tidyverse", "dplyr", "ggplot2", "factoextra", "tcltk" );

ipak(packages);

if (!require("BiocManager", quietly = TRUE)) {install.packages("BiocManager")};

BiocManager::install("MsBackendMgf");

BiocManager::install("yaml");

BiocManager::install("pcaMethods");

if (!requireNamespace("devtools", quietly = TRUE)) {install.packages("devtools")};

devtools::install_github("antonvsdata/notame"); # notame HAS to be installed this way.

Normalization steps

Normalization of IDs

In this step the IDs from mzmine are normalized by adding a method prefix and create standardized numbering filled with zeros. Valid method prefixes are RPMET, HIMET and RPLIP.

Sample Weight Normalization (optional)

This step normalizes potential differences in sample amount, e.g. weights or cell counts. A normalization factor is calculated based on a target value and the acutal sample weights. As weights either actual weights for tissues, protein amount or hoechst counts can be used.

Missing Value Imputation (per batch)

Intensity Drift Normalization (per batch)

Intensity drift normalization performed to normalize for changes within the injection sequence. The notame package is used for this stip.

Batch Normalization

QC filtering