Welcome to the Network Homologs database
Please use the side bar on the right to search for homologs of genes. Make sure to select the appropriate organism.
Select a BLAST e-value cutoff to be used as a filter. Since the scores are computed in real time increasing the number of homologs evaluated increases the processing time. The default probability cutoff used to define functional neighborhoods is 0.5. Change this value if you want to increase or decrease the neighborhood size. By default the top 50 neighbors are kept even if they fall bellow the cutoff. Uncheck the "soft cutoff" box to disable this option. One you click search you will be taken to a homolog list view.
List ViewThis view displays a list of homologs for your query gene along with their BLAST e-values (relative to the query genome) as well as the network similarity score. The left three columns display the numbers used to compute the functional similarity value. The columns titled "N1" and "N2" contain the sizes of the two neighberhoods within the relevant background set (Note: this value is not always the same for the query gene as it depends on what gene families are present in the genome of the other homolog). The last value is the size of the intersection and it is also the link to Pair View.
This is a detailed view of gene families that were present in the neighborhood overlap and contribute to the similarity score. Each cluster is displayed as the complete list of genes present in it for the model organisms used in our study. Bellow the lists of genes is the enrichment for this set of gene families. A family is considered annotated to a particular process if any of the member genes have been annotated to that process via experimental evidence.
Cluster view provides a global view of the functional relationships between multiple homologs. To cluster genes check off the desired genes in List View (more than 20 will not display well), you can make use of the "Select All" or "Select Range" buttons and click on "Submit Selected". This will take you to Cluster View. The top graphic is a hierarchical clustering of the genes selected using functional similarity as a distance metric with complete linkage. The bottom graphic is a matrix view of all the annotations that were enriched among the families present in the neighberhoods of the individual genes. In many cases clusters in the top view will correspond to simlar patterns of enrichment.