| Popis: |
Alzheimer’s disease (AD) is a chronic, advanced brain sickness disease that slowly destroys memory and thinking skills and, in the end, the ability to perform routine tasks. This disease is caused by the abnormal clumping of proteins such as amyloids around the brain cells. The identification of proteins involved in Alzheimer’s is essential to understand the disease and to discover and design the drugs. Experimental processes involving in-vitro or in-vivo experiments for this purpose are very time-consuming, laborious, and highly costly. However, costly and tedious experimental procedures can be performed efficiently by targeting the most probable proteins involved in Alzheimer’s predicted and ranked through a computational method with better generalization accuracy. In this study, we have proposed a machine learning (ML)–based predictive model to identify proteins potentially involved in Alzheimer’s. Through a series of simulation studies, we have shown that our proposed model by using protein sequence information only gives state-of-the-art generalization performance with an area under the precision-recall curve of 0.93 verified through various ML-centric and biologically relevant techniques and metrics. Through data mining in this study, we have also performed feature analysis to identify the role of individual amino acids in such proteins. Python code for feature extraction, training, and evaluating our proposed models together with the dataset is available at the URL: https://sourceforge.net/projects/alzheimer-associated-proteins/files/. |
Plný text