Databases Comparison

A new database specialized in rare kidney diseases has been developed and will be compared with MONDO. This new database aims to improve the diagnosis of patients with these conditions.

To compare the database performance, three diagnostic tools were used: Phenomizer [1], Phrank [2], and Phen2Disease [3]. By comparing the diagnostic quality of real patients using these tools, it can be determined with statistical relevance which database is better. The tools were reimplemented according to their respective papers for simplicity. The implementation is presented in this code, with unit tests to prove the validity of the code.

The semantic measures library [4] was used for the implementation of the tools.

Description

This code is designed to test the validity of the implementation of the three diagnostic tools.

VSCode is recommended for using the code. The main file for the tests is DiagnosisToolsTest.java located in the folder kidney/src/test/java/com/phenoold/kidney. You can either run the main() function or run the tests independently.

References

1. Köhler, S. et al. Clinical diagnostics in human genetics with semantic similarity searches in ontologies, 85, 457. URL.
2. Jagadeesh, K. A. et al. Phrank measures phenotype sets similarity to greatly improve Mendelian diagnostic disease prioritization, 21, 464–470. URL. Publisher: Nature Publishing Group.
3. Zhai, W., Huang, X., Shen, N. & Zhu, S. Phen2disease: a phenotype-driven model for disease and gene prioritization by bidirectional maximum matching semantic similarities, 24, bbad172. URL.
4. Harispe, S., Ranwez, S., Janaqi, S. & Montmain, J. The semantic measures library and toolkit: fast computation of semantic similarity and relatedness using biomedical ontologies, 30, 740–742. URL.