Supplementary:

Statistical identification of gene association by CID in application of constructing ER regulatory network

• R Programs
  assoc_tests.r
  
  
  • This R program provides one example of computing the statistics and
    permutation p-values for one test gene, CCNA2, based on 8 methods
    discussed in the manuscript.
    
    
  • The outputs, including 8 statistics and their p-values for CCNA2, will show
    on the R console after the program is sucessfully excuted.
    
    
  • The permutation p-value would be slightly altered due to its nature of randomness.
    
    
  • To run the R code, copy all commands in the file and paste them on R console.
    R is a freely available language and environment for statistical computing and
    graphics. It can be downloaded from http://www.r-project.org/
    
    
• Microarray data had been deposited to GEO database with accession number GSE9309. Only part of samples in GSE9309 were used for this study. They were GSM237139, GSM237140, GSM237141, GSM237143, GSM237144, GSM237148, GSM237149, GSM237150, GSM237151, GSM237152, GSM237153, GSM237154, GSM237155, GSM237156, GSM237157, GSM237158, GSM237159, GSM237161, GSM237165, GSM237166, GSM237167, GSM237168, GSM237174, GSM237175, GSM237176, GSM237177, GSM237178, GSM237179, GSM237180, GSM237181, GSM237182, GSM237183, GSM237184, GSM237185, GSM237186, GSM237187, GSM237188, GSM237190, GSM237191, GSM237193, GSM237194, GSM237195, GSM237197, GSM237198, GSM237199, GSM237200, GSM237202, and GSM237203.
  
  We also note that the rank-invariant linear LOWESS is the same with the rank consistent LOWESS, which is the official term applied by Agilent.