You are here: Home Services Bioinformatics Tools Visualizing genome-wide data

Visualizing genome-wide data

The complexity of the information stored in databases and publications, and the growing number of data coming from high throughput platforms, make difficult to contrast experimental results with the existing information. When this results are to be explored at genomic level, or consist of a huge number of data points (44k CGH array), the task seems impossible.

Furthermore, existing genomic information sources are dispersed and are formatted in many different ways. When performing this kind of genomic analysis, we need to adapt or develop new bioinformatics tools able to process the data, to infer new information from the different sources, and also to show all the information in a easy and clear way.

To explore the genomic features associated with DNA replication time (RT) observations obtained with aCGH, we implemented a computational pipeline able to gather information coming from Ensembl, UCSC and data published in literature. The workflow dynamically calculates variable size windows (depending on the density of the source data), centered around each array probe, and estimates the value of each corresponding genomic feature in each particular window. This is done thousands of times for each different genomic feature under study. A further step implies correlating the computed data with the experimental values. To perform this kind of analysis, we optimized our computation resources to compare millions of annotations per hour.

All our computational inferred relations should then be presented in an user-understandable and easily searchable way. To accomplish this task, we adapted one of the most flexible genomic browsers available, Gbrowse.


Gbrowse allows the researchers to browse the RT segmented regions along the chromosomes, together with the other computed genomic features, which permits to to study different relations at microarray probe level. Following this approach, we can now look for significant associations between the experimental data and the genomic features, and incorporate new annotations at will.