Project Description

This is the repository to redo all analysis done within the Maternal Obesity project. (Manuscript Link will be added as soon as possible). If you want to:

• redo the analysis for one omic type, please go to the respective subdirectory and execute the main.R script
• check results tables regarding each Omic (DE analysis, ORA-analysis, CoCena) please checkout Database.html
• reuse custom functions, checkout the utils- directory

Usage

Run main analysis

To run the main analysis and produce all plots shown in the manuscript and supplementary you need to:

1. Open R
2. Set the working directory as the Folder that includes all analysis
3. Run main.R (manually or source it)

Note: If you source the file - current settings within the file might limit the output to the requested option. To change you need to change respective settings (Marked in the scripts)

Example Lipidomics:

setwd("MaternalObesity")
source("lipidome_analysis/main.R")

This will output the Heatmap of all Lipid-Classes for wt and the subset for ko.

CoCena

The CoCena was conducted following closely the provided example Notebook. Therefore, to rerun the analysis one has to open the respective .Rmd Files. In here, the code chunks can be either all executed at once or, again, manually. To execute all, in RStudio, there should appear in the top left corner of the Editor a button "Run". Select here "Run All".

Data Overview

Data Repositories

The raw data with their respective Metadata can be found under the following addresses:

• Transcriptomics: https://www.ncbi.nlm.nih.gov/geo/query/acc.cgi?acc=GSE237408
• Lipidomics: http://dx.doi.org/10.21228/M81D9R
• Metabolomics: http://dx.doi.org/10.21228/M81D9R

Processed Data

All data needed to run the analysis can be found in the data directory. Note, this might not necassarily be the raw data. For transcriptomics the fastq files were summarized to a .dds object using the provided script transcriptomics_analysis/CreatingDDS.R. For Lipidomics and Metabolomics, data was respectively parsed to three data matrices including the entite x samples matrices as well as one for sample Metadata and one for entitie Metadata. This format can be easily parsed to a SummarizedExperiment- object, which again can be reused in a plethora of Bioconductor Workflows.

Database File

If you are just interested in more analysis-results, for example in exact statistical figures for a gene of interest you can rely on the Database.html file to check and also easily download respective data. An html file can be opened in any browser. Please be patient, it will take a couple of minutes to load the file completely. On the left of the file you can find a dynamic look up table as soon as you start scrolling to jump to respective section. In this File you can find the data-tables concerning:

• Transcriptomics
  • DE analysis of Kupffer Cells and respective ORA analysis of the DE Sets (Up and Down regulated)
  • hCoCena identified cluster-gene sets and their respective ORA analysis ofr Kupffer Cells, Hepatocytes wt and ko
• Metabolomics ANOVA results for all metabolites
• Lipidomics
  • Log Fold Changes actual numbers (no scaling, as done for visualisation purposes) for wt and ko
  • their respective actual p-values for wt and ko

Further Information

If you need any help or have further questions about this repository please drop me an email! [email protected] or [email protected]