✨ TL;DR
This study develops a machine learning model using molecular descriptors to screen over 7,000 natural compounds for potential anti-Alzheimer activity, identifying 73 promising candidates. The cheminformatics approach demonstrates moderate predictive performance and highlights key molecular features associated with therapeutic potential.
Alzheimer disease is a progressive neurodegenerative disorder characterized by amyloid-beta plaques and tau protein tangles that cause neuronal death, leading to severe cognitive decline and loss of independence. Despite being the most common cause of dementia in older adults, no definitive cure exists, and the exact etiology remains unclear, though age, genetics, lifestyle, and cardiovascular health are contributing factors. There is a critical need for new therapeutic compounds, and traditional drug discovery methods are time-consuming and expensive, making computational screening approaches valuable for identifying potential candidates from large compound libraries.
The researchers developed a cheminformatics-based predictive model functioning as a drug screening system for anti-Alzheimer compounds. They collected over 7,000 natural medicinal compounds from three databases (ChEBI, SynSysNet, and INDOFINE) and preprocessed them using Open Babel software. Molecular descriptors were calculated using Dragon software to characterize the chemical properties of each compound. A Random Forest classifier was trained on known approved Alzheimer treatments to learn patterns associated with therapeutic efficacy, then applied to screen the large compound library for potential anti-Alzheimer activity based on molecular features including atomic polarizability, bond multiplicity, and non-hydrogen bond counts.