Welcome to the web CePa page.

Web CePa is a tool to find significant pathways from diffrential gene expression data. Different from the traditional over-representation analysis (ORA) methods that find significant pathways without the topological information, CePa takes into account the pathway structure information so that it can capture new findings that closely related to the current biological problems.

Have a try


Web CePa has the following advantages:

Other versions

Implemented as an R package

How to use

Users are suggested to register and log in to get the full use of the tool. Although anonymous users can also use the tool, their results would be deleted as soon as their browser sessions expire.

The data input form is illustrated in figure 1. Users should specify their gene list, background gene list and the pathway catalogue. The gene ID type should be specified because the tool will transform gene IDs to gene symbol which is the working identifier in the tool. As an option, our tool provides several pre-defined background gene lists that support some boradly used microarray platforms.

Figure 1. Data input form

After submitting your data, a step that check your data would be processed. The check program would print several simple statistics of your job (illustrated in figure 2). If there are errors, users are not allowed to go to the next step.

Figure 2. Check jobs

Once the input data is valid, the job would be sent to a job system. If many jobs are running, the current job would be queued and waiting. The job would be executed as soon as the sever has time to handle your job. Figure 3 illustrates the job system and the summary of the results.

Figure 3. Job system and the result of the analysis

Centrality measurements

Six centrality measurements are selected by default.

Data source

Pathway catalogues

Pathway catalogues are collected from Pathway Interaction Database.

Supported gene ID types

Pre-defined background gene lists


Software implementation

The web service is hosted by Apache2.2 and data is administrated by MySQL. Perl is used to build the web service. R is used to do the calculation. The calculation part is processed through a paralelle computing environment in several computing servers. The R packages for parallele computing are snow and multicore.

Personal information we collect

The user's password is encrypted before inserted into database, which means, we cannot see user's real password.


Web CePa is developed and maintained by Zuguang Gu. Users are welcome to contact jokergoo[at]gmail.com.