Titolo della tesi: Machine learning methods applied to classify complex diseases using genomic data
Complex diseases present challenges in disease prediction due to their multifactorial nature. In this work, I explored the prediction of four different complex diseases, multiple sclerosis, Alzheimer’s disease, schizophrenia, and Parkinson’s disease using machine learning methods. The primary objective of this research is to investigate the robustness and variability of machine learning models constructed using genomic data in the context of predicting complex diseases. Different models will be tested to classify affected and healthy individuals, and their performance will be compared with the results obtained using polygenic risk score. The secondary goal is to apply explainability methods to extract the features considered more informative by the models. This is because understanding which genomic variants are considered informative for disease discrimination during the training process can provide significant insights into the underlying genetic basis of the diseases and identify potential targets for further investigation.