1. ‘Omics’ - Analysis of high dimensional Data
Achim Tresch Computational Biology

2. Schedule Monday Lecture: Introduction to Omics
Motivation: Transcriptomics
Experimental techniques
Data analysis (overview)
Data exploration of univariate data
Measures of location and scale
Bar plot, box plot, histogramm, density plot
Data exploration of bivariate data
Odds ratio, correlation crosstable, scatter plot, QQ-plot
Sebastian Dümcke
Henrik Failmezger
Exercises:
Introduction to R and Bioconductor
Forensic bioinformatics
Arijit Das

3. Omics Omics (Wikipedia):
Omics informally refers to a field of study in biology such as genomics, proteomics or metabolomics. Omics aims at the collective characterization and quantification of pools of biologically / biochemically similar molecules that translate into the structure, function, and dynamics of an organism or organisms.
Ingredients for omics research:
High throughput experimental techniques for the simultaneous measurements of large numbers of molecules
Statistical methods for the appropriate analysis of high dimensional data.
Generally, Omics data analysis takes longer than data generation!

4. low-medium throughput
Genomics: Transcriptomics
Techniques for RNA quantification
Northern Blot
Reporter genes
Reverse Transkriptase PCR
Microarrays
RNA-Sequencing
low-medium throughput
high throughput

5. Northern Blot
RNA (or DNA) is separated by the size on a gel, transfered to the membrane and hybridized with gene-specific probe
RNA -> Nothern blot
DNA -> Southern blot
Low throughput and poor quantification
Molecular Biology of the Cell (© Garland Science 2008)

6. Reverse transcription
RT-PCR
Reverse transcription
RNA
DNA
PCR
The course of PCR (amount of double-stranded DNA) is monitored using a specific fluorescent dye
Differences in concentration of particular mRNA in different samples can be calculated as 2N, with N being the difference in the number of cycles to obtain the same amount of product
Medium throughput, high precision N
Molecular Biology of the Cell (© Garland Science 2008)

7. Microarrays
mRNA is converted to cDNA and labeled, and subsequently hybridized to an array of gene-specific probes (either spotted cDNA samples or oligonucleotides, either one or two sample(s) per microarray)
Differences in expression between samples are determined as a ratio of fluorescence signals at individual spots.
High throughput,
medium precision (low dynamic range)
Molecular Biology of the Cell (© Garland Science 2008)

8. Zyklusvorlesung Molekularbiologie WS 2009/10
Next generation sequencing (NGS)
Massively parallel sequencing techniques enable sequencing of genome-wide cellular RNA pools
Typical sequencing read lentgh is nucleotides  RNA or cDNA has to be fragmented
A single run comprises reactions, depending on a platform, so most RNAs are covered by multiple “reads“  read occurence for a particular gene reflects expression level
High throughput,
precision depends on sequencing depth (#reads)
Zyklusvorlesung Molekularbiologie WS 2009/10

9. Next generation sequencing (NGS)
Illumina (Solexa) sequencing
DNA fragments are coupled to glass slide and subjected to Bridge amplification.
individual reads of bp are produced at a time by using fluorescently labeled removable terminator tags
Sample preparation
Sequencing

10. Transcriptomics with Microarrays
Workflow of a microarray experiment
Experimental design
Technical
performance
Data
mining
Statistical
analysis
Frame a biological
question
Obtain the samples
Extract fluorescence
intensities
Cluster analysis
and pattern recognition
Isolate total RNA
Normalize data to
remove biases
Choose a microarray
platform
Study lists of
genome ontologies
Label cDNA or mRNA
Estimate expression changes
Search for regulatory motifs
Perform the
hybridizations
Decide on biological
and technical replicates
Reconstruct regulatory circuits
Identify differentially expressd transcripts
Scan the chips
Design the series
of hybridization
Design validation
and follow-up experiments
days
weeks
days-weeks
months
After: Gibson, G and SV Muse, 2004.

11. Transcriptomics with Microarrays
labeled sample
Sample amplification and labeling
sample injected into microarray
RNA sample
Probe array hybridization
Fluorescence intensity translated into mRNA abundance
Probe array scanning and intensity quantitation
Probe array washing and staining

12. RNA Sample preparation

13. RNA Sample preparation

14. Hybridization onto microarray
Quakenbush, 2006

15. Hybridization onto microarray

16. Hybridization onto microarray
mismatch probes
perfect match probes
probe pair
Each gene is represented by probe pairs of 25nt length, consisting of a perfect match probe and a mismatch probe.
Perfect match probes are complementary to specific sequences of the target gene, preferentially located at the 3’ end of a gene.
The mismatch probe is identical to the perfect match probe, except for the middle base. It is designed to detect unspecific binding.

17. Affymetrix Microarrays – Probe Synthesis
For the extension of all oligonucleotides by one base, four litographic steps with complementary masks are performed, one mask for each base A, C, T, G.

18. Affymetrix raw Data Greyscale and false color image
of the fluorescence readout

19. Detection of differentially expressed genes
Data Analysis
Detection of differentially expressed genes
Identification of similar samples and co-regulated genes in a multi-sample comparison
genes
samples

20. Data Analysis Cluster Analysis Pearson correlation matrix Venn diagram
Summary statistics up-/down regulation
[ Phenotypic analysis ]
Koschubs et al., EMBO J, 2009.

21. Response to chemical stimulus Vitamin metabolic process
Data Analysis: Gene Ontology
Response to chemical stimulus
Vitamin metabolic process

22. Microarray Databases ArrayExpress (European Bioinformatics Institute)
Gene Expression Omnibus (NCBI)

23. Dietmar Martin Gene Center,
Acknowledgement
Dietmar Martin Gene Center,
LMU Munich