Back to index

Chapter 5: Representing the results

Once the search and anyalysis is done, Biblio-MetReS will give a message that you need to acknowledge before the results are displayed. The initial display of the resuts is shown in Figure 10.

The first pane is this new view shows the documents that were found during the search. You can scroll down or up this list to view the title for the different documents, which you can sort accord to the different sorting options. If you click the "View document in the web browser" link, the document will open in your default browser.

figure 10

Figure 10: Output: List of analyzed documents

If you click the "View genes that are found in the document" link, a dialog will open (Figure 11). the first box in the dialog will show you the list of genes detected in the document. The second box will show you the gene pairs found in sentences within the document. The final box will show you the gene pairs found within the different paragraphs of the document. To continue using the application and to view other results you should close this dialog.

figure 11

Figure 11: Dialog showing the genes and gene pairs found within one document

The second pane in the results main view will display the list of genes that was found during the analysis. If you scroll the list and double click any of the genes, the corresponding KEGG gene page will open in your default browser. There , you will links to corresponding gene pages in others daabases and to the pathways you gene may be involved in (Figure 12).

figure 12

Figure 12: The gene list view

the third pane in the results main view provides information about the co-occurrence of the different genes in tabular form (Figure 13). You can export all this information as a text gile with tab-delimited fields. You can then open this file in any text editor or import it to a program like excel. You can sort the results by ascending or descending orde in each column, by clicking on the column tag. The first column shows you the list of all gene pairs that was found in the analysis. The second column provides the mutual information associated with this gene pair. The third column provides the p-value for estimating how likely it is that this association is random. The lower the p-value is, lower likelihood is. The fourth column indicates the number of time that the gene pair has been found, These columns provide numbers for the analysis of co-occurring genes in sentences. If gene pair does not co-occur in any sentence, all numbers will be 0. The fifth columm lets you visualize the sentences in which the gene pair was found, if any.

figure 13

Figure 13: Tabular view of Biblio-MetReS results

The sixth column shows the mutual information calculated for co-occurrende in paragraph, the seventh column shows the p-value calculated for the co-occurrence in paragraph, and the eigth colum shows the number of documents in which gene pair was found, The last column allows you to visualize the documents in which the gene pair was found.

In adition to the information descibed above, Biblio-MetReS gives the possibility of analyzing the results in graphical from In teh graphs, the nodes represent the genes that were found during the search and analysis of documents. These genes will be united through edges. If an edge joins two nodes, these two genes have been found to co-occur in the same document. The thickness of the edge is porportional to the number of co-occurences between the two genes in the set of analyzed documents, while the color is proportional to the p-value. Red predicts low significance, while green predicts high significace. By default the graphs are circular. However, you can click an drag nodes with your mouse in order to personalize the appearance of the according to your preferences. Furthermore, by double/clicking on a nodes you can rename that node by typing the new name with your keyboard.

The graph titled "Results of Biblio-MetReS: Interaction Between Genes in Paragaphs" (Figure 14) shows the co-occurrences found between genes in the same paragaphs in all processed document.

figure 14

Figure 14: Interaction graph for gene/proteins co-occurring in the same paragraphs within the documents analyzed while running the application.

If you press the "View sentence interaction graph" button, this will open a graph titled "Results of Biblio-MetReS: Interaction Between Genes in Sentences". This show the co-occurrences found between different genes in the same sentence in all processed document (Figure 15). Even though the default graph is circular, you can edit the graph and drag nodes around in order to identify meaningful biological modules.

figure 15

Figure 15: Interacton graph for genes/proteins co-occurring in same sentences within the documents analyzed while running the aplication

Results from this graph are expected to be more indicative of biological interaction between genes than those of the previous graph.

Once you are done with the analysis, just click "New Search" button or exit the application.


< Previous chapter Next chapter >