Gene Example¶

The get_equivalent_features() method is used to find genes equivalent to the input/target feature.

from graphkb.match import get_equivalent_features

genes = get_equivalent_features(graphkb_conn, 'KRAS')

This will use a similar algorithm to what we have seen above in the disease matching example.

initial graph

In the graph above the relationship types shown are: GeneralizationOf (G), ElementOf (E), DeprecatedBy (D), and CrossReferenceOf (X).

Match by Name¶

As before, the first thing done is to match the input name

match by name

Resolve Aliases¶

The next step is to resolve equivalent names of the current set of terms.

resolve aliases

Follow the Elements Tree¶

The next step is to follow the element relationships. This is treated the same as the subclassing except now our "tree edge" is the ElementOf relationship type.

element tree

Resolve Final Aliases¶

Finally we expand the current set of terms by alias terms again to capture aliases of the more general parent and more specific child terms expanded in the previous step.

resolve aliases

We have now collected all of the different terms for KRAS