Sarah Carratt 4.05.11

From OpenWetWare
Revision as of 10:05, 5 April 2011 by Sarah Carratt (talk | contribs)
Jump to navigationJump to search

Review of Homework

Timepoints

  • Schade = 0m, 10m, 30m, 120m, 12h, 60h
  • wt and Dgln3 = 15, 30, 60, 90, 120

Resaon for Statistics

  • Is this computed ratio a true difference or a fluxuation due to noise?
  • T statistic to P value
  • Things that are part of the equation...
    • size of effect, aka, is it a big difference?
    • amount of variation
    • number of replicates
      • Schade (confusing)
      • wt (4,5,4,5,5)
      • Dgln3 (4,4,4,4,4)

P Values

  • more than 300 p values significant by chance = part of problem, use as guide only
  • multiple testing problem corrected by:
    • Bonferroni corrections
    • Comparing fold changes and p.values

Most Significantly Changed Genes

  • NSR1 = commonly accepted

Moving Forward

Hierachical Clustering

  • Each row is gene
  • Each column is chip
  • If a group of genes is all on or off, maybe regulated by same transciption factor...why? Logic of the cell says cell needs all turned on at once because they are all part of same process.

Gene Ontology (GO)

  • Description of genes, think of this as a process of categorizing genes
  • Yeast, flies, mouse, ect. databases combined to organize biological vocabulary so they are consistant in papers, get rid of excess synonyms.
  • Written definitions of terms in 3 categories.
    • Biological process
    • Molecular function
    • Cellular component
  • Genes are asssigned to terms by model organism databases