Translational Bioinformatics

Application
(CSE-896)

Omics Data Integration

Spring 2025
Lecture 09
Middle Integration:
Most commonly
used

Uses ML models for consolidating data without concatenation

of features before analysis
1
Major Challenges in Integration:
• Some of the major challenges that must be tackled for reliable
integration of omics datasets include:

1. Data heterogeneity

2
Major Challenges in Integration:
• Some of the major challenges that must be tackled for reliable
integration of omics datasets include:

1. Data heterogeneity

Whole-genome vs Transcriptome data

3
Major Challenges in Integration:
• Some of the major challenges that must be tackled for reliable
integration of omics datasets include:

2. Uneven and missing data

Variations in the number of features/samples

Features may be absent in certain samples

4
Major Challenges in Integration:
• Some of the major challenges that must be tackled for reliable
integration of omics datasets include:

3. High dimensionality

Large p, small n problem

5
Major Challenges in Integration:
• Some of the major challenges that must be tackled for reliable
integration of omics datasets include:

4. Computational Performance

6
ML Methods for Multiomic Integration:
• mixOmics:

mixOmics is an R package for exploring and integrating omics data, including transcriptomics,
proteomics, lipidomics, microbiome, metagenomics and beyond. The mixOmics package includes tools
for data integration, biomarker discovery, and data visualization.

7
ML Methods for Multiomic Integration:

R and Rstudio were set up …

mixOmics was installed …

8
ML Methods for Multiomic Integration:

Reproducibility is very important while performing data analysis

9
Setting up R/RStudio:

10
ML Methods for Multiomic Integration:
• Reproducibility means …

• That the methods of an experiment can be repeated?

• The results of subsequent methods based on those methods

would generate identical results?

• If two groups were analyzing the same data, they would reach the
same conclusions?

11
ML Methods for Multiomic Integration:

“Reproducibility is a minimum necessary condition

for a finding to be believable and informative.”
Goodman et al., 2016

12
Setting up R/RStudio:

13
Setting up R/RStudio:

14
ML Methods for Multiomic Integration:
• Ensuring reproducibility … Data Organization

• Make plans for appropriate storage of raw and processed data

• Create project directories

• Stay organized during the analysis

You can't have any sort of reproducibility

without good data management

Adapted from Goldman, June 2020 and Goldman and Obrycki, December 2020 15
Setting up R/RStudio:

https://phdcomics.com/comics.php?f=1689 16
ML Methods for Multiomic Integration:
• Ensuring reproducibility … How to?

• Document everything

• Create READMEs that detail data organization, analysis methods,

dates, naming conventions, etc.

• Version and parameters that were for the analysis tools

• What were the exact commands that you ran throughout the analysis?

• Annotate code with comments

Adopted from Goldman, June 2020 17

ML Methods for Multiomic Integration:
• Ensuring reproducibility … Tools for documentation

• Documentation … RMarkdown, Jupyter Notebook

• Version Control … Git, Bitbucket

• Collaboration and Version Control … GitHub, Bitbucket

• Containerization … Docker

18
Basic R and RStudio:

R is not just a programming language

… but an environment

19
Basic R and RStudio:

Available here: https://r4ds.hadley.nz/

20
Basic R and RStudio:

RStudio is a freely available open-source IDE

Runs on all major platforms: Windows, Mac OS, UNIX/Linux.

21
Basic R and RStudio:
• Let’s create our first project directory using RStudio.

• Open RStudio

• Go to the File menu and select New Project.

• In the New Project window, choose New Directory. Then, choose New
Project. Name your new directory Intro-to-R and then “Create the
project as subdirectory of:” the Desktop (or location of your choice).

• Click on Create Project.

22
Basic R and RStudio:
• Let’s create our first project directory using RStudio.

23
Basic R and RStudio:
• Go to the File menu and select New File, and select R Script.
• Go to the File menu and select Save As..., type Intro-to-R.R and
select Save

24
Basic R and RStudio:

3. History/
1. Code Editor Environment

4. Plots/ Help/
Packages
2. Console

25
Basic R and RStudio:

What is a project in RStudio?

26
Basic R and RStudio:
• A directory that contains everything related to your analyses for
a specific project.

.Rproj file is created

… keeps track of command history and variables in the environment

… can be used to reopen the project in its current state

27
Basic R and RStudio:
• When a project is (re) opened within RStudio the following actions are
taken:
• A new R session (process) is started

• The .RData file in the project’s main directory is loaded, populating the environment with
any objects that were present when the project was closed

• The .Rhistory file in the project’s main directory is loaded into the RStudio History pane
(and used for Console Up/Down arrow command history).

• The current working directory is set to the project directory.

• Previously edited source documents are restored into editor tabs

• Other RStudio settings (e.g. active tabs, splitter positions, etc.) are restored to where they
were the last time the project was closed.
28
Basic R and RStudio:
• Setting Up Working Directory:

• Use the getwd() function

29
Basic R and RStudio:
• Setting Up Working Directory:

• Use the getwd() function

Changing working directory

30
Basic R and RStudio:
• Structuring your working directory:
For instance, separate directories
for raw & processed data

Create three directories:

data
results
figures

31