Lecture 17: Microarrays to analyze gene expression

Outline:


Some notes on cell differentiation (http://en.wikipedia.org/wiki/Cellular_differentiation)

Types of cells:
  • Adult stem cells 
    • Self-renewal -- can divide but maintain undifferentiated state
    • Multipotency - can generate daughter cells of may different cell types
  • Germ cells
    • Give rise to gametes (eggs and sperm)
    • Undergo meiosis (for reproduction)
  • Somatic cells (cells that are not stem cells or germ cells)
Differentiation:
  • Process by which a cell becomes more specialized
  • Doesn't change the DNA
  • Changes a cell's size, shape, membrane potential, metabolic activity, signal response
  • Switches from one pattern of gene expression to another
  • Cell signaling by growth factors
  • Several mechanisms for epigentic regulation via DNA methylation
Which cells end up as which is a random process:

DNA Microarrays 
Measure expression level (mRNA) of large numbers of genes simultaneously 

Several technologies: gene-chip technology (Affymetrix)



Microarray applications:
  • Find out which genes are up-regulated or down-regulated between two or more conditions (gives genes that are candidates for causing the variations in condition)
Example: Compare average expression of genes in samples from tumor and non tumor and determine which are significantly different (t-test or Mann-Whitney test). The t-test assumes the distributions are normal, while tests such as Mann-Whitney to not.   
  • Find out how much variation is due to subjects, how much to procedural issues, and how much to disease
Example: Use ANOVA to find out the amount of variance accounted for by each factor
  • Find out subclasses of tumors 
Example: Cluster gene expression profiles for tumors and then compare differences across clusters to look for biomarkers (e.g., Triple Negative in breast cancer).
  • Discover new pathways
Example: Look at whether GO terms and other metadata are enriched in clusters, then postulate pathways and mechanisms and test in the laboratory

There are many other types of microarrays: http://en.wikipedia.org/wiki/Microarray



Microarray analysis in R (some preliminaries)

Preliminaries:
source("http://bioconductor.org/biocLite.R")
biocLite()
biocLite("GEOmetadb")library(GEOmetadb)
if(!file.exists('GEOmetadb.sqlite')) {
getSQLiteFile()
}