Problem about detailed content of testset

[Nero-gitg]

Hi, I used the contents of the 9 files under data/disease_files on this GitHub repo to restore the disease and drug information from the PrimeKG dataset (kg.csv, nodes.csv, disease_features.csv). However, the resulting statistical information differs from the information in the paper. The number of diseases is not much different, but the number of drugs for indication and contraindications varies significantly. I am currently conducting some experimental comparisons of methods on this dataset. To ensure rigor and fairness, I would like to ask if you could provide the complete disease and drug information for these nine test set files.

Below is the statistic in your paper:

This is the my statistical result:
adrenal_gland.csv：Number of diseases 6 , Number of indications 39 , Number of contraindications 307
anemia.csv：Number of diseases 17 , Number of indications 73 , Number of contraindications 550
autoimmune.csv：Number of diseases 15 , Number of indications 88 , Number of contraindications 330
cardiovascular.csv：Number of diseases 103 , Number of indications 367 , Number of contraindications 3417
cell_proliferation.csv：Number of diseases 192 , Number of indications 1118 , Number of contraindications 1125
diabetes.csv：Number of diseases 3 , Number of indications 104 , Number of contraindications 367
mental_health.csv：Number of diseases 54 , Number of indications 341 , Number of contraindications 1318
metabolic_disorder.csv：Number of diseases 42 , Number of indications 96 , Number of contraindications 548
neurodigenerative.csv：Number of diseases 16 , Number of indications 135 , Number of contraindications 146

Looking forward to your answer!