Edge Function for Time Course Data Analysis

Introduction

Microarray technique has provided an efficient way to study gene expression at the genome level. One application of time-course microarray data is to group together genes with similar expression patterns. Genes clustered in the same group are considered to have the potential of conducting similar functions. A number of measures of similarity have been used, including correlation coefficient. While these methods focus on global features, the edge function introduced here focuses on local features. In this study, pairs of known transcription regulations are taken as functional related genes, and each pair is scored using correlation coefficient and edge function methods.

Collection of known transcription regulations in S. cerevisiae

Keyword searching using "regul* on YPD database as of Feb. 2000 yielded 1007 genes. By reviewing the published literature on these 1007 genes, 888 transcriptional regualtions were collected, of which 647 were activators and 241 were inhibitions. Altogether, 486 genes were involved in these regulations.

These genes were mapped to two time series data sets, generated from cdc28 and alpha factor synchronization.

• cdc28 data set
• alpha data set

The mapping result is shown as follows:

	time points	time interval	gene mapped	activations	inhibitions
cdc28	17	10 min	366	469	155
alpha	18	7 min	335	343	96

Evaluation of known regulations using correlation coefficient and edge function methods

regulations sorted by alphabet

• 469 activations mapped in cdc28 data set
  • 1 - 469 (very slow, recommend to use segments)
  • 1 - 50
  • 51 - 100
  • 101 - 150
  • 151 -200
  • 201 - 250
  • 251 - 300
  • 301 - 350
  • 351 - 400
  • 401 - 469
  
  
• 155 inhibitions mapped in cdc28 data set
  • 1 - 155 (slow)
  • 1 - 50
  • 51 - 100
  • 101 - 155
  
  
• 343 activations mapped in alpha data set
  • 1 - 343 (slow)
  • 1 - 50
  • 51 - 100
  • 101 - 150
  • 151 -200
  • 201 - 250
  • 251 - 300
  • 301 - 343
  
  
• 96 inhibitions mapped in alpha data set
  • 1 - 96
  • 1 - 50
  • 51 - 96
  
  

regulations sorted by correlation coefficient

• 469 activations mapped in cdc28 data set
  • 1 - 469 (very slow, recommend to use segments)
  • 1 - 50
  • 51 - 100
  • 101 - 150
  • 151 -200
  • 201 - 250
  • 251 - 300
  • 301 - 350
  • 351 - 400
  • 401 - 469
  
  
• 155 inhibitions mapped in cdc28 data set
  • 1 - 155 (slow)
  • 1 - 50
  • 51 - 100
  • 101 - 155
  
  
• 343 activations mapped in alpha data set
  • 1 - 343 (slow)
  • 1 - 50
  • 51 - 100
  • 101 - 150
  • 151 -200
  • 201 - 250
  • 251 - 300
  • 301 - 343
  
  
• 96 inhibitions mapped in alpha data set
  • 1 - 96
  • 1 - 50
  • 51 - 96
  
  

regulations sorted by edge function

• 469 activations mapped in cdc28 data set
  • 1 - 469 (very slow, recommend to use segments)
  • 1 - 50
  • 51 - 100
  • 101 - 150
  • 151 -200
  • 201 - 250
  • 251 - 300
  • 301 - 350
  • 351 - 400
  • 401 - 469
  
  
• 155 inhibitions mapped in cdc28 data set
  • 1 - 155 (slow)
  • 1 - 50
  • 51 - 100
  • 101 - 155
  
  
• 343 activations mapped in alpha data set
  • 1 - 343 (slow)
  • 1 - 50
  • 51 - 100
  • 101 - 150
  • 151 -200
  • 201 - 250
  • 251 - 300
  • 301 - 343
  
  
• 96 inhibitions mapped in alpha data set
  • 1 - 96
  • 1 - 50
  • 51 - 96
  
  

Description of edge function

	click here