Rice Coexpression Analysis
The Pearson correlation coefficient of two genes is calculated based on 1,081 rice Affymetrix microarray data (NCBI GEO AC: GPL2025) after filtering low quality slides. On the rice whole genome scale, there are 37,993 genes which have Affymetrix probeset matched, of which 34,016 have unique Affymetrix probeset available and only these genes were included in this database. So total 578,527,120 correlation coefficients for every pair of selected genes were obtained.
In order to build the coexpression network, a PCC cutoff is needed to consider the pair of genes coexpressed. To choose an appropriate PCC cutoff , we examined the changes in the node number, edge number, and network density using different cutoffs. The following figure shows that as the cutoff value increases, both the node number and edge number decrease; however, as the cutoff reaches a relatively high value, the decreasing rate of edges becomes slower than that of nodes, which might lead to an increase in the network density. Indeed, the network density reaches minimum around PCC 0.75 and increases thereafter. So PCC 0.75 was selected for the coexpression analysis. The resulting coexpression network contains 8,521 nodes and 496,682 edges, and a network density of 0.01368. These coexpressed gene pairs are the base of this database.
There are two main functions in this coexpression section. One is the coexpression analysis which will give out the coexpressed genes of query gene under a certain PCC cutoff. The other one is coexpression network construction of query genes under certain PCC cutoff and depth of network.
The following figure shows the correlation coefficient distribution on the whole genome scale and the interval is 0.01.
Last modified: Friday, 30-Mar-2012 08:13:16 Pacific Daylight Time
| Available Data: |
| Platforms: 6 |
| Experiments: 105 |
| Hybridations: 1,867 |
Last Update: Apr. 2012