1 of 15

Integrating transcriptome and DNA methylation data using iNETgrate, an examples multiomics analysis.

19 Oct 2023�Hands-on in the Neurepiomics Course

Dr. Habil Zare, PhD

Assistant Professor at the Bigss Institute

PI of the Oncinfo Lab

2 of 15

2

Central dogma (20th century)

  1. Genome (DNA sequence)
  2. Transcriptome (gene expression)
  3. Proteome

​

​

​

3 of 15

3

Omics data (21st century)

  • Genome (DNA sequence)
  • Transcriptome (gene expression)
  • Proteome
  • Epigenome (methylome & histone modifications)
  • Metabolome
  • Lipidome
  • Microbiome
  • ...

​

4 of 15

4

The true power of multiomics

  • “Most multiomics approaches use cross-platform omics data to validate one another when investigating a specific phenotype. This approach is useful but does not exploit the full potential of the multiomics data.”
  • Disagreements between the different levels of molecular information can reveal the mechanisms that explain these inconsistencies e.g., overexpressed transcripts corresponding to reduced protein levels.

5 of 15

5

Network analysis

Rationale: Interaction between thousands of genes in hundreds of samples can be modeled in a single network.

6 of 15

6

What are the nodes in the network?

Network of genes

Network of patients

7 of 15

7

SNFtool: Similarity network fusion

​

8 of 15

8

SNFtool: Similarity network fusion

​

  • Patient clustering, survival analysis, etc. :)
  • Little biological interpretation :(

9 of 15

9

Each node is a gene

Pearson correlation

10 of 15

10

Previous hands-on

  • Genome (DNA sequence)
  • Transcriptome (gene expression)
  • Proteome
  • Epigenome (methylome & histone modifications)
  • Metabolome
  • Lipidome
  • Microbiome
  • ...

​

Transcriptome (gene expression)

Gene network analysis

based on level of expression of thousands of genes

11 of 15

11

Pigengene (Neurepiomics 2019)

Survival analysis

12 of 15

12

Pigengene is on Bioconductor

if (!requireNamespace("BiocManager", quietly = TRUE))

install.packages("BiocManager")

​

# The following initializes usage of Bioc devel

BiocManager::install(version='devel')

​

BiocManager::install("Pigengene")

  • Download the source package (Version >1.11.34).
  • Follow the pdf and code chunks in Pigengene/inst/doc.

13 of 15

13

Update: integrating DNA methylation

  • Genome (DNA sequence)
  • Transcriptome (gene expression)
  • Proteome
  • Epigenome (m
  • Metabolome
  • Lipidome
  • Microbiome
  • ...

​

Transcriptome (gene expression)

DNA methylation

iNETgrate:

iNETgrate

14 of 15

14

iNETgrate is on Bioconductor

if (!require("BiocManager", quietly = TRUE))

install.packages("BiocManager")

# The following initializes usage of Bioc devel

BiocManager::install(version='devel')

BiocManager::install("iNETgrate")

15 of 15

15

Future work

​

?

Genomic variance

Metabolome data

Incorporate other types of data into the network analysis.