FAQ Examples

1. Some genes seem to be double-annotated in Tree View output of Onto-Express. (Back to FAQ)

Q. In the tree view results of Onto-Express here, it can be  observed that in total 17 locus ids were contained within the main term Antioxidant activity. 7 of these IDs were no further sub classified. However, 4 of these 7 locus ids were also contained within the sub term peroxidase activity.
Why are these 4 locus ids not removed from the main term and only included within the sub term, because the remaining 8 locus ids from the peroxidase activity are also not included within the main term antioxidant activity. In the current output these 4 locus ids seem to be double-annotated.

Fig. 1 Sample Tree View output

A. For the given example, to get the p-value for peroxidase activity, the nodes should be expanded such that all genes are seen in that category and use the p-value associated with peroxidase. To get the p-value for the more general antioxidant activity, the nodes should be collapsed to re-group all the genes, which will eliminate any duplicates and calculate the appropriate p-value. In the expanded view 7 + 1 + 12 +1 genes can be seen. When  the antioxidant activity is collapsed, a smaller number equal to the number of unique genes 7+1+12+1 = 21 minus the number of  redundant ones (4) = 17 will be seen.

 

2. There are some differences between the results from matlab using "hygepdf" function and the results generated by Onto-Express.

Q. While using U133+ chip and input 148 unique genes, and concerning the GO:0005524, whose corresponding parameters are as follows:

Unique Input Total: 26

Unique Reference Total: 2465

Unique UniGene Total: 26

Unique Reference UniGene Total: 1046

By using hygepdf(26, 54120,148, 2465), the result is 2.6623e-009 and the result given by the Onto-Express is 5.79E-10. Using hygepdf(26,38572,148,1046), the result is 3.2856e-014.

A. First of all, the correct function to use in MATLAB to calculate the p-value is, hygecdf(X,M,K,N) (not hygepdf).
Secondly, the parameters that are passed to hygepdf function are incorrect. MATLAB defines hygecdf function as p = hygecdf(X,M,K,N), where p is the probability of drawing upto X of a possible K items in N drawings without replacements from a group of M objects. So in case of this example, X is 26, M is 54644 (for HGU133 PLUS 2.0 array), K is 2465 and N is 148. Hence, your MATLAB function call should be hygecdf(26, 54644, 2465, 148). This call in MATLAB returns p = 1.0. This means that the number of genes in your given category is so unusually high that the probability of having a value lower than this just by chance is basically 1.
Onto-Express returns 5.79E-10. The p-value of 1.0 indicates that it is in the right-hand side tail of the distribution. Since it is on the right-hand side of the distribution, and in order to be consistent (low values should always mean high significance) Onto-Express subtracts it from 1.0, hence p-values = 0.0. The reason that you get p-value = 5.79E-10 instead of 0.0 is due to the inherent floating point error in any computer. The color used by Onto-Express tells you that this is in the right tail of the pdf, ie. more than expected.  The p-value close to 0 tells you that this is highly significant and extremely unlikely to occur just by chance.

3. How to increase the the Java runtime memory for the Java Plug-in?

A.  Follow the steps below:

1. Open up the Java Plug-in from Control Panel window.

For Windows, click on start -> control panel -> Java Plug-in.

For Linux/Unix, type the following command in a terminal

$ <JAVA installation directory>/jre/bin/ControlPanel

OR

$ <JRE installation directory>/bin/ControlPanel

Fig. 2a Java Plug-in Control Panel Windows

2. Click on the "Advanced" tab of the Java Plug-in window. 

Fig. 3 Java Plug-in window Advanced tab

3. There will be a form at the bottom of this tab labeled "Java Runtime Parameter". Specify the minimum and maximum memory that you want to allocate here by adding the following line

-Xms128m -Xmx256m

This will assign the minimum memory of 128 MB and maximum memory of 256 MB to the Java Plug-in.

Note: User can play around with the amount of memory she wants to use, depending upon the size of the Flat View output gif image  she needs to save. The maximum memory we could assign to Java Plug-in on a Windows machine with 2000 MB RAM was 846 MB.

4. Click on "apply" button and close the Java Plug-in window.

5. Close all your browser windows and restart the browser for the new settings to take effect.

Note: If the browser crashes after you restart it, try reducing the minimum and maximum memory specified in step 3 and try again.


 

 

Webmaster - ISBL Team

   

copyright 2003 Intelligent Systems and Bioinformatics Laboratory, Computer Science Department, Wayne State University