• Project Background

  • About Cancer

  • Identifying Cancer Drivers

• Project

  • Mutation Analysis in Cancer

  • Building a Driver Predictor

• Disease with a bad name:
  • prevalent:
    • 14 million new incidents/year ,
  • Cause of death:
    • 8 million deaths/year,
    • 1/5 of worldwide mortality
  • Related to everyone:
    • 28% ~ 41% individual lifetime risk of developing cancer

• Cancer is
  • Group of cells with limitless growth potential
    
  • Caused by epigenetic or genetic changes
• It's a complex disease because of:
  • Cancer genes > 600
    
  • Heterogeneity
    
  • Genome instability:
    • a few cancer "drivers" and dorminant "passengers"
  • and…

… > 100 subtypes (maybe more)

Precision and prevention:

• diagnosis
  • imaging test / biopsy -> molecular biomarkers / sequencing
  • diagnosis -> early screening
• treatment:
  • radiotherapy chemotherapy ->
    • targeted therapy
    • immunetherapy

More details revealed by advanced methods:

• Cancer epigenetics
• Tumor Microenvironment
• Single-cell Sequencing
• Cancer genomics

One of the main goal is to

Definition:

Two types of cancer genes:

1. POGs
   • Activation by mutation is advantageous to cancer
2. TSGs
   • Inactivation by mutation is advantageous to cancer

• positive selection signals
  • MutsigCV, MUSiC, Oncodrive
  • Compare gene mutation burden with a background mutation model
  • mutation Clustering in sequence / protein structure

• Gene or protein Interaction based analysis
  • NetBox, HotNet2, DriverNet
  • neighbor-based forcasting

• Copy number variation and Gene expression level
  • OncodriveCIS, S2N, Conexic
  • abnormal elevated / supressed gene expression

• Definition:

Mutations that contribute to cell growth advantage when altered

• somatic and germline

• Types:
  • Coding region:
    • Single nucleotide variation
      
    • Insertion / deletion
      
  • Noncoding region
    
  • Copy Number Variation / Aneuploidy

Two main strategies:

• Mutation recurrence among patients
  • fail to spot less frequent drivers
• Functional impact of mutations
  • SIFT, Polyphen, CHASM, FATHMM, etc
  • strategies:
    • DNA conservation
    • Clustering of mutations
    • Affecting functional sites

• For driver Genes:
  • A lack of "gold standard" driver genes
    
  • Unable to "discover" all known driver genes
• For driver mutations:
  • lack of "gold standard"

• First step of driver discovery: understand them.

• Goal: analyze cancer mutations and their correlation with DNA and protein structural features

• We hypothesize that driver is enriched in recurrent mutations, next few slides test this hypothesis