Skip to main content

Table 2 Number of features in each category and information resource for the feature extraction.

From: CpGIMethPred: computational model for predicting methylation status of CpG islands in human genome

Category

# Features

Resource

General attributes

3

Gardiner-Garden criteria [19], obtained from UCSC Genome Browser

DNA sequence composition

tetramer frequency

256

calculated by in-house code based on definition

 

tetramer z-score

256

calculated by in-house code based on formula (1)-(3)

Conserved TFBS's/elements

conserved TFBS's

230

calculated by in-house code based on UCSC information [24]

 

conserved elements

2

calculated by in-house code based on conserved elements [15] from UCSC

Structural properties

DNA 3-D conformation

6

calculated by in-house code based on formula [25]

 

nucleosome positioning propensity

4

calculated by in-house code using nucleosome organization map [27]

Functional roles of nearby genes

2

calculated by in-house code for enrichment analysis

Histone modifications

histone methylation

46

calculated by in-house code based on the data set from [31]

 

histone acetylation

36

calculated by in-house code based on the data set from [33]