Mission Study

• Start Date: Aug 02, 2005
• Research Drive Directory: \Massion\050802HJ
• Data Type: Micro Array

Data information

• Total Features (rows): 2464(original); 2415(average)
  • Note: need to filter out with more than 30% missing data
• Total Samples (cols): 75
  • smoking 1 group: 30
  • smoking 2 group: 22
  • smoking 3 group: 23
  • adeno: 42
  • squamous: 31
• Original Data: \Massion\050802HJ\original

Output

• Score Summary, Winner Summary, Top100, Accuracy, Cluster: \Massion\050802HJ\delivery

Analysis Requests

• Top100
• Cutoff
• Group Info
  • Smoking1 (30) vs Smoking2 (22)
  • Smoking1 (30) vs Smoking3 (23)
  • Smoking2 (22) vs Smoking3 (23)
  • Adeno (42) vs Squamous (31)
• train-test: randomly select 70% as Training and 30% as Testing
• Look for array CGH pattern that is associated with DRC, BICS, SEX and FAMCAN_FDR(0 vs >0), breaking it down by the 3 Smoking status Margaret.

Redo at 01/16/2006

• Data from Aixiang: after average, after log2, no missing value.
  • Total Features (rows): 2415
  • Total Samples (cols): 75
• In group Smoking3, Never Smokers (17) vs Long Term Quitters (6)