Identification of Novel Inhibitors and Biomarkers for Alzheimers Disease using Computational Systems Polypharmacology and Machine Learning Approach

Abstract

The current research is focused on the management of the pathophysiology of the newlineAlzheimer s Disease (AD). The AD is a neurological irreversible disease characterized by the newlineabnormal accumulation of amyloid beta and neurofibrillary tangles in the brain. Although newlinethese hallmarks can identify in the later stages. Currently used drugs for AD can only newlineslowdown the progression but they cannot halt the AD progression. We have analysed newlinethousands of AD genes and proposed the biomarker identification method as well as gt 1 newlinemillion compounds were screened against several potential targets followed by 5 µS newlinemolecular dynamics simulation (MDS) and lead compounds were proposed. Firstly, we have newlineretrieved the AD genes from the DisGeNet dataset and 13,504 features were calculated. newlineThese features were evaluated by using 16 machine learning (ML) methods. The result newlineshowed that network-based features are showing and#8764;92% accuracy while sequence-based newlinefeatures only showing and#8764;52% accuracy. The feature selection approach increases and#8764;2-3% newlineaccuracy. Best performing features were used for the feature fusion analysis. By utilizing newlinefeature fusion approach, we have constructed 24 new features with 6,020 dimensions. The newlinecomposition of k spaced amino acid pairs (CKSAAP) based fused features showed and#8764;10% newlineincrease in the accuracy. Then 8 CKSAAP fused features with alone CKSAAP were newlineconsidered for hyperparameter tuning where we have not seen gt=70% accuracy for any newlinefeature. Therefore, we left the sequence-based features and used network-based features for newlinethe hyperparameter tuning approach where we have seen that network-based features are able newlineto classify between the AD and non-AD genes with 97.23 and 96.55% accuracy for training newlineand test dataset respectively. Then proposed model was validated by using the blind dataset newlineobtained from AlzGene and Gene Expression Omnibus databases. Finally, we have proposed newlineAlzGenPred method as a standalone tool to the scientific community. newlineThe GSK3and#946; and CDK5 are two key enzymes which majorly phosphorylate the tau newlineprotein.