Stepwise connection: a focus on the INE algorithm

This notebook provides the steps to build a network starting from a list of genes, and using the Iterative Neighbor Expansion (INE) algorithm.

%%time
from neko.core.network import Network
from neko._visual.visualize_network import NetworkVisualizer
from neko.inputs import Universe, signor
from neko._annotations.gene_ontology import Ontology
import omnipath as op
import pandas as pd

CPU times: user 15 µs, sys: 4 µs, total: 19 µs
Wall time: 21 µs

1. Define the list of genes of interest

genes = ["PIK3CA","MAP2K1","GSK3B","MAPK14","CTNNB1","AKT1","MAP3K7"]

output_nodes_prosurvival = ['CCND1','MYC','TCF7L2']

2. Specify SIGNOR resource

resources = Universe()
resources = signor("../neko/_data/signor_db.tsv")  # this function accept only tab separated values
resources.build()

3. Create a network using as input the gene set.

3A. Import genes as network nodes

new_net = Network(initial_nodes = genes, resources=resources.interactions)

3B. Connect nodes

The INE algorith connects the input nodes by looking and introducing all their immediate neighbors into the network. If the nodes remain unconnected, the immediate neighbors of the newly introduced nodes are added into the network. The process ends either when a complete network is reached or until the user-specified threshold of expansion steps (i.e., max_len) is specified.

The algorithm also allows the definition of outputs, which are nodes that should be connected to the network with only incoming edges. For instance, these nodes might represent phenotypic markers (e.g., proliferation) of interest.

At the end of the algorithm, any nodes that do not have a source in the edge dataframe and are not in the output node are removed..

%%time
new_net.connect_as_atopo(max_len=1, strategy="radial",outputs=output_nodes_prosurvival)

CPU times: user 7.83 s, sys: 2.45 ms, total: 7.83 s
Wall time: 7.84 s

new_net.edges

	source	target	Type	Effect	References
0	P31749	P49840	None	inhibition	11035810
1	P31749	P46937	None	inhibition	12535517
2	P31749	P42345	None	stimulation	15829723
3	P31749	P04150	None	inhibition	24291004
4	P31749	Q01860	None	stimulation	23041284
...	...	...	...	...	...
190	SIGNOR-PF4	P01106	None	stimulation	32482868
213	Q16539	P24385	None	inhibition	8702807
229	P28482	P01106	None	stimulation	8386367
241	P49841	P24385	None	inhibition	16504004; 9832503; 23552696
242	P49841	P01106	None	inhibition	11018017; 16023596; 14563837

150 rows × 5 columns

visualizer = NetworkVisualizer(new_net, color_by='effect', noi=True)
visualizer.render()

visualizer = NetworkVisualizer(new_net, color_by='effect', noi=True, predefined_node="CCND1")
visualizer.render()

visualizer = NetworkVisualizer(new_net, color_by='effect', noi=True, predefined_node="MYC")
visualizer.render()

visualizer = NetworkVisualizer(new_net, color_by='effect', noi=True, predefined_node="TCF7L2")
visualizer.render()