Get node weight for a simple bipartite graph

Question

I've created a bipartite networkx graph from a CSV file that maps Disorders to Symptoms. So, a disorder may be linked to one or more Symptoms.

for disorder, symptoms in csv_dictionary.items():
    for i in range (0, len(symptoms)):
        G.add_edge(disorder, symptoms[i])

What I need is to find what Symptoms are connected to multiple diseases and sort them according to their weight. Any suggestions?

Answer 1

You can use degree of the created graph. Every symptom with degree larger than 1 belongs to at least two diseases:

I've added some example csv_dictionary (please supply it in your next question as minimal reproducible example) and created a set of all symptoms during the creation of the graph. You could also think about adding these information as node feature to the graph.

import networkx as nx

csv_dictionary = {"a": ["A"], "b": ["B"], "c": ["A", "C"], "d": ["D"], "e": ["E", "B"], "f":["F"], "g":["F"], "h":["F"]}

G = nx.Graph()

all_symptoms = set()
for disorder, symptoms in csv_dictionary.items():
    for i in range (0, len(symptoms)):
        G.add_edge(disorder, symptoms[i])

        all_symptoms.add(symptoms[i])

symptoms_with_multiple_diseases = [symptom for symptom in all_symptoms if G.degree(symptom) > 1]
print(symptoms_with_multiple_diseases)
# ['B', 'F', 'A']

sorted_symptoms = list(sorted(symptoms_with_multiple_diseases, key= lambda symptom: G.degree(symptom)))
print(sorted_symptoms)
# ['B', 'A', 'F']