COMPUTATIONAL ANALYSIS OF MULTILEVEL OMICS DATA FOR THE ELUCIDATION OF MOLECULAR MECHANISMS OF CANCER Presented by Azeez Ayomide Fatai Supervisor: Junaid.

COMPUTATIONAL ANALYSIS OF MULTILEVEL OMICS DATA FOR THE ELUCIDATION OF MOLECULAR MECHANISMS OF CANCER

Presented by

Azeez Ayomide FataiSupervisor: Junaid Gamieldien

Note: You only have 10-15 minutes maximum, so I suggest presenting only an introduction + section 2

INTRODUCTIONPre-genomic era

Cloning genes at the site of proviral integrationFunctional assaysPositional cloning

Post-genomic era

High-throughput technologiesWES and NGSDNA methylationGenomic hybrization Copy number alteration Gene expression profiling DNA methylation

Simultaneous study on a cohort of samplesUnderlying mechanismsPrognostic and predictive biomarkersTarget identification

Cancer genomics project & Databases TCGAICGC

Tools in clinicMammaPrintOncotype DXBreast cancer profiling test (HOXB13/IL17RB)

Breakdown of my study

1. Network-based identification of candidate cancer genes• Identification of functionally relevant genes in copy

number regions• Co-expression and transcriptional analysis

2. Identification of differentially expressed miRNAs and their target genes in the GBM network

3. Identification of prognostic miRNAs for progression-free survival prediction

4. Identification of prognostic protein coding transcripts? genes for progression-free survival prediction

5. Pathway-based and machine learning based feature selection (describe more completely)

Identification of differentially expressed miRNAs and their targets in the GBM network

INTRODUCTION

• Discuss the aims and objectives and the rationale of this section here

• State your hypothesis

Flowchart for miRNA analysis in GBM

Materials and Methods

• Add a slide that gives specific details of the method used to identify differentially expressed miRNAs (and WHY they were chosen)

• R modules• Underlying statistical tests• p-value cutoffs• fold-change cutoffs (if any)• Describe the samples – numbers, classes, etc• etc

Differentially expressed miRNAs between tumour and non-neoplastic brain samples

Is there any way to rank these and then list only the ‘best’?

Also, be careful to explain what the red text is highlighting

Convert the underxpressed fold change as follows: -1/fold-change

- that will make 0.1 = -10 fold change for example

…continues

Underexpressed miRNA-overexpressed gene network

Produce a better layout if possible – Also highlight any known cancer related miRNAs and genes

Very important: stress that the agreement between miRNA and mRNA expression direction illustrate that the experimental data (and conclusions) are trustworthy

Any known important genes thatyou can point out to the audience?

Overexpressed miRNA-underexpressed gene network

Highlight any known cancer related miRNAs and genes.

Also, are there any miRNAs that appear to be regulatory ‘hubs’ based on number of genes they interact with? If so, point them out.

Pathways enriched with miRNA target genes

Discussion

• What did you learn from this section?• Find anything important?• Eg. is there any disregulated miRNA that looks

like it plays dominant major role?• Can it be a drug target?• Is there any gene that can be a drug target?• Etc

Conclusions

• Biological take home message (e.g. miRNA-mRNA networks play a role in GBM… etc)

• Mention what you took from this chapter into the next chapters and just give a BRIEF verbal description of the predictive features you found (just to show again that this is just part of a bigger study)

Acknowledgements

• Your university that sponsors your PhD

• Anyone other than me that helped you with data or analysis or tips/clues even in the smallest way

• Etc