Result: Radni okvir Mendelove randomizacije u programskom jeziku Python ; Mendelian Randomization Framework in Python Programming Language

Title:
Radni okvir Mendelove randomizacije u programskom jeziku Python ; Mendelian Randomization Framework in Python Programming Language
Contributors:
Vladimir, Klemo
Publisher Information:
Sveučilište u Zagrebu. Fakultet elektrotehnike i računarstva.
University of Zagreb. Faculty of Electrical Engineering and Computing.
Publication Year:
2025
Collection:
Croatian Digital Theses Repository (National and University Library in Zagreb)
Document Type:
Dissertation/ Thesis master thesis
File Description:
application/pdf
Language:
Croatian
Rights:
http://rightsstatements.org/vocab/InC/1.0/ ; info:eu-repo/semantics/openAccess
Accession Number:
edsbas.3554D16E
Database:
BASE

Further Information

U ovom radu obrađena je metoda Mendelove randomizacije (MR), statistički pristup koji koristi genetske varijante kao instrumentalne varijable za ispitivanje uzročnih odnosa između izloženosti i zdravstvenih ishoda. Prikazana je teorijska pozadina metode, osnovne pretpostavke koje moraju biti zadovoljene, te osnovne i napredne metode analize.U praktičnom dijelu analiziran je TwoSampleMR paket programskog jezika R koji koristi podatke iz baze OpenGWAS. Na temelju tog sustava izrađena je vlastita Python implementacija koristeći biblioteke pandas, numpy i ieugwaspy. Sustav je modularno organiziran i obuhvaća osnovne funkcionalnosti poput dohvata instrumenata i ishoda, harmonizacije podataka, izvođenja MR analiza, testova validacije i vizualizacije rezultata.Evaluacija implementacije provedena je usporedbom rezultata između R i Python sustava na nekoliko primjera. Dobiveni rezultati pokazali su visoku podudarnost i potvrdili ispravnost implementiranog okvira. ; This thesis explores the method of Mendelian Randomization (MR), a statistical approach that uses genetic variants as instrumental variables to investigate causal relationships between exposures and health outcomes. The theoretical background of the method is presented, including the key assumptions that must be satisfied, as well as both basic and advanced analytical techniques.In the practical part, the TwoSampleMR package from the R programming language, which utilizes data from the OpenGWAS database, is analyzed. Based on this system, a custom Python implementation was developed using the pandas, numpy, and ieugwaspy libraries. The system is organized in a modular fashion and includes core functionalities such as instrument and outcome data retrieval, data harmonization, execution of MR methods, validation tests, and result visualization.The implementation was evaluated by comparing the results between the R and Python systems across several examples. The obtained results showed high concordance and confirmed the correctness of the implemented framework.