Abstract
Abstract
The progressive, chronic nature of Alzheimer's disease (AD), a form of dementia, defaces the adulthood of elderly individuals. The pathogenesis of the condition is primarily unascertained, turning the treatment efficacy more arduous. Therefore, understanding the genetic etiology of AD is essential to identifying targeted therapeutics. This study aimed to use machine-learning techniques of expressed genes in patients with AD to identify potential biomarkers that can be used for future therapy. The dataset is accessed from the Gene Expression Omnibus (GEO) database (Accession Number: GSE36980). The subgroups (AD blood samples from frontal, hippocampal, and temporal regions) are individually investigated against non-AD models. Prioritized gene cluster analyses are conducted with the STRING database. The candidate gene biomarkers were trained with various supervised machine-learning (ML) classification algorithms. The interpretation of the model prediction is perpetrated with explainable artificial intelligence (AI) techniques. This experiment revealed 34, 60, and 28 genes as target biomarkers of AD mapped from the frontal, hippocampal, and temporal regions. It is identified ORAI2 as a shared biomarker in all three areas strongly associated with AD's progression. The pathway analysis showed that STIM1 and TRPC3 are strongly associated with ORAI2. We found three hub genes, TPI1, STIM1, and TRPC3, in the network of the ORAI2 gene that might be involved in the molecular pathogenesis of AD. Naive Bayes classified the samples of different groups by fivefold cross-validation with 100% accuracy. AI and ML are promising tools in identifying disease-associated genes that will advance the field of targeted therapeutics against genetic diseases.
Publisher
Springer Science and Business Media LLC
Subject
Cellular and Molecular Neuroscience,Neurology (clinical),Biochemistry
Reference45 articles.
1. Al-Thani HF, Ahmad MN, Younes S, Zayed H (2021) Genetic variants associated with alzheimer disease in the 22 arab countries: A systematic review. Alzheimer Dis Assoc Disord 35(2):178–186
2. Antoniadi AM, Du Y, Guendouz Y, Wei L, Mazo C, Becker BA, Mooney C (2021) Current challenges and future opportunities for XAI in machine learning-based clinical decision support systems: a systematic review. Appl Sci 11(11):5088
3. Bagabir, S., Ibrahim, N. K., Bagabir, H., & Ateeq, R. (2022). Covid-19 and artificial intelligence: Genome sequencing, drug development and vaccine discovery. J Infect Publ Health
4. Barrett T, Wilhite SE, Ledoux P, Evangelista C, Kim IF, Tomashevsky M, ... Soboleva A (2012) NCBI GEO: archive for functional genomics data sets—update. Nucl Acids Res 41(D1):D991–D995
5. Bhattacharyya R, Teves CAF, Long A, Hofert M, Tanzi RE (2022) The neuronal-specific isoform of BIN1 regulates β-secretase cleavage of APP and Aβ generation in a RIN3-dependent manner. Sci Rep 12(1):1–12
Cited by
2 articles.
订阅此论文施引文献
订阅此论文施引文献,注册后可以免费订阅5篇论文的施引文献,订阅后可以查看论文全部施引文献