Heteroscedastic Personalized Regression Unveils Genetic Basis of Alzheimer’s Disease Stratified by Cognitive Level
Motivation: The presence of heterogeneity among patients is a common phenomenon for certain diseases, including Alzheimer's disease, cancer, and COVID-19. Regrettably, conventional models that share numerous parameters across all specimens struggle to discern this heterogeneity and identify the influential factors for individuals. To tackle this challenge, it is of paramount importance to devise a personalized model adept at capturing the inter-patient heterogeneity inherent in such instances.
Result: In this work, we propose an innovative approach called Heteroscedastic Personalized Regression (Het-PR) to estimate the heterogeneity across samples and obtain personalized models for each sample. We demonstrate the effectiveness of Het-PR through both simulation and real data experiments. In the simulation experiment, we show that Het-PR outperforms other state of art models in capturing inter-sample heterogeneity. In the real data experiment, we apply Het-PR to Alzheimer's data, and show that it can identify persuasive selected factors for each individual patient. Software for Het-PR is available in github.com/rong-hash/Het-PR.