Fig. 6
Symptom-centered diagnostic disease modules. (A) Distribution of LCC z-score for the largest connected components formed by related diseases of 105 symptoms. (B) Distribution of network separation (Sab) for all symptom pairs, with an average network separation < Sab> very close to 0, indicating a high degree of similarity between some symptoms of psychosomatic disorders. (C) The network distance (Dab) between symptom pairs and the co-occurrence (Co-diseases count) between symptom pairs are negatively correlated (Pearson’s correlation = -0.63). (D) There is a negative correlation between the relative risk (RR) of symptom pairs and the network distance Dab (Pearson’s correlation = -0.29), confirming that shorter network distances between symptoms can predict their co-occurrence. (E,F) Visualization of the disease modules associated with the symptoms “Disorientation” and “depression”.
