Hindawi Computational and Mathematical Methods in Medicine Volume 2021, Article ID 6329041, 8 pages https://doi.org/10.1155/2021/6329041

Research Article Identification of the Hub in Alzheimer’s

Huiwen Gui,1 Qi Gong,1 Jun Jiang,2 Mei Liu ,3 and Huanyin Li 1

1Department of Neurology, Minhang Branch, Zhongshan Hospital, Fudan University, 170 Xinsong Road, 201199 Shanghai, China 2State Key Laboratory of Genetic Engineering, School of Life Sciences, Fudan University, Shanghai 200438, China 3Department of General Practice, Minhang Branch, Zhongshan Hospital, Fudan University, 170 Xinsong Road, 201199 Shanghai, China

Correspondence should be addressed to Mei Liu; [email protected] and Huanyin Li; [email protected]

Received 6 May 2021; Accepted 3 July 2021; Published 16 July 2021

Academic Editor: Tao Huang

Copyright © 2021 Huiwen Gui et al. This is an open access article distributed under the Creative Commons Attribution License, which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly cited.

Purpose. Alzheimer’s disease (AD) is considered to be the most common neurodegenerative disease and also one of the major fatal affecting the elderly, thus bringing a huge burden to society. Therefore, identifying AD-related hub genes is extremely important for developing novel strategies against AD. Materials and Methods. Here, we extracted the expression profile GSE63061 from the National Center for Biotechnology Information (NCBI) GEO database. Once the unverified gene chip was removed, we standardized the microarray data after quality control. We utilized the Limma software package to screen the differentially expressed genes (DEGs). We conducted Gene Ontology (GO) and Kyoto Encyclopedia of Genes and Genomes (KEGG) analyses of DEGs. Subsequently, we constructed a -protein interaction (PPI) network using the STRING database. Result. We screened 2169 DEGs, comprising 1313 DEGs with upregulation and 856 DEGs with downregulation. Functional enrichment analysis showed that the response of immune, the degranulation of neutrophils, lysosome, and the differentiation of osteoclast were greatly enriched in DEGs with upregulation; peptide biosynthetic process, translation, ribosome, and oxidative phosphorylation were dramatically enriched in DEGs with downregulation. 379 nodes and 1149 PPI edges were demonstrated in the PPI network constructed by upregulated DEGs; 202 nodes and 1963 PPI edges were shown in the PPI network constructed by downregulated DEGs. Four hub genes, including GAPDH, RHOA, RPS29, and RPS27A, were identified to be the newly produced candidates involved in AD pathology. Conclusion. GAPDH, RHOA, RPS29, and RPS27A are expected to be key candidates for AD progression. The results of this study can provide comprehensive insight into understanding AD’s pathogenesis and potential new therapeutic targets.

1. Background festations of AD patients include the declined cognitive function, mental and behavioral disorders, and decreased Alzheimer’s disease (AD) is typical hippocampal amnesia capability of daily living [5]. AD is classified into three stages and cognitive disorder [1]. It is characterized by amyloid pla- in view of the deteriorated degree of cognitive capability and ques (extracellular), neurofibrillary tangles (intracellular), physical function [6]. AD devastates numerous people and and structural and functional changes in memory-related has become a chief medical and social burden worldwide [7]. regions [2, 3]. There are about 50 million people with As known to all, a variety of complex pathogenic factors, dementia around the world and about 10 million newly such as genetic and environmental factors, lead to the occur- emerged diseases annually; 60-70% of these cases are patients rence of AD [8, 9]. Up to date, it is yet elusive towards the with AD. It is shown that the number of people suffering from mechanisms involved in AD’s pathogenesis, and efficient dementia around the world has increased twofold more from methods are incomplete to prevent and treat AD [10]. 1990 to 2016. This trend is mainly attributed to the aging Though several clinical treatments have been applied in com- and growth of the population [4]. Due to its slow or invisible bating the cognitive and behavioral deficits associated with onset, it is hard to be conscious of its initial. The main mani- AD, they are still needed to be improved due to limitations. 2 Computational and Mathematical Methods in Medicine

4

2

0

−2

−4

Figure 1: Heat map analysis of identified DEGs between AD specimens and healthy controls. The genes with upregulation and downregulation were indicated as red and green, respectively.

Gene chip technology is a toolset that arranges a large rence and development of Kawasaki disease (KD) [15], which number of nucleic acid molecules in a large-scale array on a provides a new idea for the prevention and treatment of KD small carrier and detects the strength of the hybridization sig- coronary artery expansion. nal by hybridizing with a labeled sample and then determines This article is dedicated to screening and identifying the the number of detected molecules in the sample [11]. It has differentially expressed genes (DEGs) in the the advantages of high sensitivity and accuracy, quickness profile GSE63061 and DEGs related to AD. We performed and simplicity, and the ability to detect multiple diseases at function and pathway enrichment analysis of the DEG and the same time [12]. The past decade has witnessed the discov- subsequently constructed the protein-protein interaction ery and validation of more than a dozen of risk genes related (PPI) network. Finally, we obtained several genes related to to AD. In human prostate , the use of gene chip tech- AD: GAPDH, RHOA, RPS29, and RPS27A. nology to explore the role of GAB2 in human prostate cancer cells provides a new therapeutic target for prostate cancer 2. Materials and Methods [13]. Using lncRNA microarray gene chip technology, it was found that AC002454.1 has a significantly high expres- 2.1. Extraction of Microarray Data. The gene expression pro- sion in children with acute leukemia [14], which is related file GSE63061 on Illumina HumanHT-12 V4.0 expression to the immunophenotype and prognosis of children with beadchip was acquired from the Gene Expression Omnibus acute leukemia to a certain extent. Through gene chip tech- (GEO) of NCBI (http://www.ncbi.nlm.nih.gov/gds/) [16]. A nology, it has been identified that miR-937 in peripheral total of 112 samples, comprising 72 AD samples and 40 con- blood mononuclear cells (PBMCs) is involved in the occur- trol samples, were studied. Computational and Mathematical Methods in Medicine 3

Biological process −log10 (P.value)

Neutrophil activation involved in immune response 20

Neutrophil degranulation

Neutrophil mediated immunity

Phosphorylation 15

Cytokine−mediated signaling pathway

Cellular response to cytokine stimulus

Antigen processing and presentation of peptide antigen via MHC class I 10

Innate immune response activating surface receptor signaling pathway

Transmembrane receptor protein tyrosine kinase signaling pathway

Negative regulation of protein serine/threonine kinase activity 5

0.10 0.15 0.20 0.25 0.30

Gene ratio Count 25 50 75 (a) P KEGG analysis −log10 ( .value)

Lysosome 12

Osteoclast diferentiation

Phagosome

Tuberculosis 10

Infuenza A

Shigellosis

Acute myeloid leukemia 8

Apoptosis

Vibrio cholerae infection

Bacterial invasion of epithelial cells 6

0.20 0.24 0.28

Gene ratio

Count 15 25 20 30 (b)

Figure 2: Biological process (BP) and KEGG analysis for the DEGs with upregulation. (a) BP analysis and (b) KEGG pathway analysis. 4 Computational and Mathematical Methods in Medicine

P Biological Process −log10 ( .value)

Peptide biosynthetic process 35

Translation

SRP−dependent cotranslational protein targeting to membrane

Cotranslational protein targeting to membrane

Protein targeting to ER

Nuclear−transcribed mRNA catabolic process, nonsense−mediated decay

Viral gene expression

Viral transcription

Cellular macromolecule biosynthetic process

Ribosome biogenesis 30

0.2 0.3 0.4

Gene ratio Count 50 60 70 (a) P KEGG analysis −log10 ( .value)

Ribosome 40

Oxidative phosphorylation

Non−alcoholic fatty disease (NAFLD)

Parkinson disease 30

Alzheimer disease

Huntington disease

Termogenesis 20

Spliceosome

Proteasome

Cardiac muscle contraction 10

0.1 0.2 0.3 0.4

Gene ratio Count 10 40 20 50 30 60 (b)

Figure 3: Biological process (BP) and KEGG analysis for the DEGs with downregulation. (a) BP analysis and (b) KEGG pathway analysis. Computational and Mathematical Methods in Medicine 5

ARID3A TMEM87A

ANKS1A TBC1D2 SYTL3 ZNF493 SLFN12 NPEPL1 AKAP17A ZNF219 LIN28B RTP5 MBD6 TIAF1

GLYR1 PATE2 ECD METRNL JARID2 PNPT1 CRAMP1L NT5C2 SAFB2 ITPRIP PIGT RBM23 YE ATS2 PHF20L1 DDX17 DNAJC28 TRERF1 BTG1 HELZ H2AFY EMILIN2 BRPF1 RERE NR4A2 TAF1C SRSF4 GON4L KDM5B UVSSA KIAA0319L TMEM91 SLC26A6 DHX16 ACIN1 BEST1 SRRM2 HIPK2 SNRNP70 GTF3C2 OCIAD1 RPRD2 XRN2 PABPN1 MTCH1 CCNK GMEB2 CDK13 SF3B4 KCNH3 RGPD2 ACSS2 TALDO1 SLC26A8 CERS2 PABPC1L PCBP1 U2AF2 HNRNPUL1 G6PD CMIP FAM120B WDR37 RBM10 DPAGT1 PATL1 PTBP1HELZ2 POLR2A LSMEM1 MAST3 INTS3 BRICD5 PPM1B HS PA1B RXRA CREB5 SPEN ACADVL SON KMT2E TSC22D3 SF1 SIK3 SIRT2 CWC25CELF2 TXNRD1 ZSWIM8 ZC3H12A MNT TMC4 RNF44 SLC16A12 XPO6 TKT EP300 CRTC2 ACSL1 SREBF1 MKNK1 CCDC97CLK2 ATXN1 MYPOP LPP PARP4 RUSC1 CEBPB CREBBP TACC3 CHFR ZBTB16 CLCN6 HS PA1A PLOD1 CEBPD RARA FBXL18 AGTRAP MCL1 LRCH4 PTPRE BICD2 RNF213 DENND4B CFL1 IQSEC1 POTEF POLG ANKFY1 PADI4 TBC1D2B VIM OSGIN1 COL7A1 CSAD NOTCH2 DENND5A PTPN1 STIM1 CTNNA1 MAP2K1 GAPDH ATP2A2 LRSAM1 ABCA7 USP32 YWHAB STAT6 LATS2 VCL MAP3K5 ABCC5 STAT3 KIF21B IRF2BPL ABCA2 PRAM1 USP48 POTEG PFN1 GORASP1 TYK2 APH1B COPA CFLAR IKBKB STAT1 PI4KB ARHGAP24 CUX1 SEC31A NDE1 ZFHX3 P4HB DISC1 MYH9 DYNC1H1 KIAA0247 BCL6 ZNF486 UNC93B1 CSNK1G2 GAB2 IKBKG SPI1 PI4KA ATP6AP1 WAS ARAP1 IL10 GDI1 GLA TECPR2 RAP1GAP2 OSCAR TIMP2 PPP1R14B FMNL1 AP1B1 SERPINA1 RTN1 KLF6 WDR1 PRKCD CSF2RA MTMR3 PLCB2 RHOA SEC23B GNS GAS7 ADRBK1 CBL MLKL MAN2A2 NCSTNTAPBP TMEM43 ACVR1B ARHGAP1 FNBP1 RAC2 GRN TCIRG1 WDFY3 AGPAT2 AKAP13 DOCK2 HLA-F MCM8 ZFYVE26 SEMA4D GAPVD1 LCP1 MAN2B2 IST1 C5orf28 ABR TSC1 SH3KBP1 FGR NPC2 GLB1 ITGAMIGF2R CD68 LTBR ITGAX TRPM2 TAP1 MICAL1 FAM65B FGD3 CDAN1 SPG 11 KIAA0556 FHOD1 DAPP1 VCAN REPS2 CD14 BOLA2 C14orf159 PREX1 LILRB3 ARHGAP27 CD53 DAGLB ARHGAP9 GAA SYNGR1 S100A10 MYO1F LILRB2 RAB24 CD163 CYBA PARVG VNN3 CAMK2G PSAP RASA2 NOMO2PILRA BRSK1 GGCX SLC15A4 PRCP SLX1A ELMO2 ADAM8 LAMP2 CYTH1 HAVCR2 ULK1 S100A 11 RAB40C ALOX5 PLIN3 KANSL1 XYLT1 PLBD1 SMAP2 SIR PA TBC1D14 CHPF2 C7orf43 NBEAL2 BNIPL RAB 11FIP1 CIDEB MYADM ENSG00000100926 C10orf54 SELPLG TM9SF4 REM2 ADAP1 NPEPPS POFUT2 SNX27 AMY1A ZSWIM4 NDST2 CPQ CNDP2 SU LT1A1 SLC16A3 AKNA PFKFB4 TGFBI FAM156A UPK3B NUTM2F

CMTM3 ATG16L2 AKIRIN2 ITPK1 CREBRF DCAF15 ABHD15 ANKRD11 ANKRD13D NBPF1 CDC42SE1 RHBDF2 FRMD8

CMTM7 SNTB2 MAP7D1 TRPC4AP ZZEF1 PPP1R10 EFHD2 MEGF9 AQP12A TTLL3

NBPF10 TBC1D3G FAM156B ZDHHC7 FRY

FAM189B

Figure 4: The PPI networks were established by significant DEGs with upregulation, which is composed of 379 nodes and 1149 PPI edges. Nodes mean , and edges mean the interaction of proteins.

2.2. Identifying the DEGs. We utilized the Limma package in 2.4. PPI Network Analysis of DEGs. We employed the R to identify the DEGs, with an adjusted P value <0.01. For STRING database (http://string-db.org) [18], which was analyzing the DEGs in-depth, we constructed a heat map applied for the Retrieval of Interacting Genes to construct utilizing the Pheatmap package (https://cran.r-project.org/ the PPI network. To dig out AD-associated hub protein and package=pheatmap) in R. key genes, we obtained the interaction between DEGs with a total score ≥ 0:4 and then constructed a PPI network utiliz- 2.3. Functional Enrichment Analyses for DEG. We performed ing the STRING database. Finally, we conducted the Cytos- the enrichment analysis of AD-associated genes by Gene cape software to visualize the network and uncover hub Ontology (GO) and the Kyoto Encyclopedia of Genes and genes with higher degrees (connected nodes) in the PPI net- Genomes (KEGG) analyses utilizing the Database for Anno- work. These genes might have a vital role in the network. tation, Visualization and Integrated Discovery (DAVID; https://david.ncifcrf.gov/tools.jsp) [17]. GO terms containing 3. Result biological processes, molecular functions, and cellular com- ponents could commendably illustrate the biological func- 3.1. Identification of DEGs in AD. All the blood of AD tion for identified DEGs. As a public encyclopedic database, patients and healthy people from the datasets (GSE63061) KEGG comprehensively analyzed the function and biochem- was used for our research. We firstly analyzed the DEGs ical pathways of selected DEGs. If P <0:05, the result is con- between AD samples and age-matched normal samples. We sidered statistically significant. obtained 2169 genes. The genes with the most significant 6 Computational and Mathematical Methods in Medicine

REXO2

BTLA VPREB3 CRIPT

TRAPPC4

STAP1

MYLPF CD72 TIMM23 ERGIC3 DPH3 ZNF140 CCDC34 FCRL2

CD79B MS4A1

MYL6 TCEAL4 DPH5 DRG1 CCDC53

BTF3

PPHLN1 SPAG7 FCER2

C 11orf1 RSL1D1 EIF3D CD52

ZNF622 NSA2 EMG1 DLX1 THOC7 RPL21 MCTS1 EIF3E ARPC3 ENSG00000229 117 RPL17 CD3D

RPL7 RPL36AL RPS14 RPL6 RPS8 RTP4 KBTBD3 OR51S1 EEF1A1 RPF2 RPL30 RPLP0 RPS12 PSMB7 ZNF136 RPL4 VAMP8

MRPL15 RPL 11 RPS15A RPS20 FAM32A RPL31 RPS3A RPS27A PSMD4 PSMD10 COPS8 RPL12 MRPL17 MRPS7 RPS27 GPR18 RPL24 RPS29 MRPL 11 RPL39 RPS25 RAN HDDC3 RPL35 RPL35A BOLA3 EEF1B2 CKS1B RPS17 SS18L2 RPL9 RPL27 CETN2 C1QBP

MRPL51 MRPL22 CCDC167 MRPL13 SRP14 MRPL27 RBX1 C15orf57 HSP90AA1 OBFC1 TPT1 GTF2H5 PFDN5 RPA3 MRPL33 NOP10 TOMM7 H2AFZ RPMS17 SNRPD2 ERH MRPS21 ATP6V1D ZNHIT3 MRPS33 RAD54B HINT1 PRIM1 NME1-NME2 NME1 NUDT1 TMA7 COX7C COX7A2 ATP5O APEX1 OPTN

TOMM6 GPN1 ATP5E SNRPB2 UQCRH GTF2A2 GTF2B NDU FA4 PPA2 PPIAL4B DNAJA1 TTC32 NDUFS5 LSM5 NDUFB2 ATP5L SNRNP27 GPN3 ATP6V1E1

ENSA COX6A1 ATP5J DNAJC8 AIDA ATP5F1 ACADM NDUFB6 NDU FA1 NDU FA2 PRDX1 SUB1 SOD1 C14orf2 ZM AT2 EMC3 UQCRHL NGRN COX17 UBL5 CWC15 OCIAD2 C19orf43 LDHB ZCCHC17 ATP5EP2 ET FA HIGD1A TRIAP1 ISY1 CLC CTGF MDH1 TTC25 SESN3

TMEM60 COX14 APOBEC3F ITLN1 SLC25A38 UFC1 PRDX4 CYB5A FABP5 BEND5 RGS10

SELK RASGRP3 ZNF359 PDCD10

ZNF626 C 11orf48 TP53RK

MPC2

Figure 5: The PPI networks were established by significant DEGs with downregulation, which is composed of 202 nodes and 1963 PPI edges. Nodes mean proteins, and edges mean the interaction of proteins.

P values are RPL36AL, LOC100132795, NDUFA2, and so on. tide biosynthetic process, translation, ribosome, and SRP- Then, we utilized a heat map of identified DEG in the dependent cotranslational protein targeting to the membrane GSE63061 database to conduct cluster analysis (Figure 1). (Figure 3(a)). Ribosome, oxidative phosphorylation, and Genes with upregulation and genes with downregulation were nonalcoholic fatty liver disease (NAFLD) were significant shown as red and green, respectively. 1313 DEGs were indi- pathway enrichment of these DEGs with downregulation cated as genes with upregulation, and 856 DEGs were indi- (Figure 3(b)). cated as genes with downregulation. 3.3. PPI Network Analysis. We carried out a PPI network 3.2. GO and KEGG Enrichment Analyses. To obtain the func- analysis to explore the interaction and hub genes of DEGs. tion and pathway of these DEGs, we used the online tool Except for disconnected nodes in the network, 379 nodes DAVID to analyze 1313 upregulation DEGs and 856 down- (proteins) and 1149 PPI edges (interactions) were demon- regulation DEGs. Functional enrichment analysis showed strated in the PPI network constructed by upregulated DEGs that these DEGs with upregulation exhibited a significant (Figure 4). Likewise, 202 nodes (proteins) and 1963 PPI edges association with the immune response of activated neutro- (interactions) were observed in the PPI network constructed phil involvement, the degranulation of neutrophils, and the by downregulated DEGs (Figure 5). Considering the infor- immunity mediated by neutrophils (Figure 2(a)). Lysosome, mation of the STRING database, we have chosen the top osteoclast differentiation, and phagosome were significant node with a higher node degree. The number of edges was pathway enrichment of these upregulated genes (Figure 2(b)). positively correlated with the importance of their function The DEGs with downregulation were mainly related to pep- in the PPI network. The edges and nodes of the connecting Computational and Mathematical Methods in Medicine 7 lines of these genes were very dense. Among the genes with ribosome, and oxidative phosphorylation. Beck et al. pointed upregulation, GAPDH and RHOA were thought of as hub out that reducing mitochondrial oxidative phosphorylation proteins and key genes. Among the genes with downregula- could lead to defects in AD, and mitochondrial dysfunction tion, RPS29 and RPS27A are the most important. This suggests was one of the early manifestations of AD [30]. they might play an important function in AD development. This study has some limitations. First, the key DEGs need to be verified by RT-qPCR. In future studies, we will collect 4. Discussion clinical samples to verify the expression levels of key DEGs through RT-qPCR. Secondly, we plan to further explore the As an intricate and refractory neurodegenerative disease, AD mechanism of key genes in AD in the animal model. seriously affects people’s life and living quality, particularly for the elderly [19, 20]. Its pathogenesis is not yet clear, and 5. Conclusion no effective cure has been developed [21]. Thus, AD is still a major problem in human diseases. Many hypotheses In summary, we identified 2169 DEGs between AD patients explain the pathogenesis of AD, including amyloid cascades, and healthy controls. Functional enrichment analysis hyperphosphorylation, neurotransmitters, and oxidative demonstrated that DEGs with upregulation displayed a sig- stress [22, 23]. However, the root cause and optimal treat- nificant association with immune response, neutrophil ffi ment plan are still di cult to achieve. Xia et al. found out degranulation, lysosome, and osteoclast differentiation; the DEGs from the single-cell microarray data of four brain DEGs with downregulation exhibited a significant associa- ff regions a ected by AD and constructed a PPI network [24]. tion with peptide biosynthetic process, translation, ribosome, Analysis shows that the increase of oxidative stress and the and oxidative phosphorylation. Subsequently, we identified changes in neuronal lipid metabolism may be some events GAPDH, RHOA, RPS29, and RPS27A as key genes in AD that occur in the early stage of AD pathology. Lee et al. by analysis of the PPI network. The purpose of this research pointed out that p38MAPK could mediate a variety of AD- is to improve our understanding of the molecular mechanism associated events, i.e., the phosphorylation of tau, neurotox- of AD through comprehensive bioinformatics analysis and fl icity, neuroin ammation, and the dysfunction of synapsis. may give a hint of developing the treatment of AD patients. Therefore, inhibiting p38MAPK is a prospective treatment for AD [25]. Kajiwara et al. believe that the expression of -4 in microglia is related to cognitive impairment in Data Availability AD [26]. Further research on caspase-4 will be beneficial for comprehending AD’s etiology and uncovering new tar- The datasets used and/or analyzed during the current study gets for treating AD. Therefore, identifying the key genes are available from the corresponding author on reasonable involved may help to further understand the development request. of Alzheimer’s disease. We used public databases to identify and screen DEGs Conflicts of Interest and related pathways in AD through various bioinformatics methods. We identified GAPDH, RHOA, RPS29, and The authors declare that they have no conflicts of interest. RPS27A as the hub proteins and key genes of AD. Previous studies have shown that GAPDH is a regulator of cell death, Authors’ Contributions and GAPDH is involved in tumor progression and has become a new therapeutic target. Research by Hwang et al. Conception and design was carried out by Huanyin Li. showed that GAPDH-mediated mitosis eliminated defective Development of methodology was done by Huiwen Gui, mitochondria and led to , which can provide a Jun Jiang, and Qi Gong. Sample collection was done by ’ potential treatment method for the treatment of Huntington s Huiwen Gui, Mei Liu and Jun Jiang. Analysis and interpreta- disease and other neurodegenerative diseases [27]. Mirabello tion of data were conducted by Huiwen Gui and Jun Jiang. et al. pointed out that RPS29 was a constituent of the small Writing, review, and/or revision of the manuscript were per- 40S ribosomal subunit which functions essentially in rRNA formed by Huiwen Gui and Huanyin Li. Huiwen Gui and Qi processing and ribosomal biogenesis and also reported that Gong contributed equally to this work. RPS29 could cause autosomal dominant Diamond-Blackfan anemia [28]. Researchers have shown that RPS27A may be a potential target of mesenchymal stem cells in treating type 2 References diabetes mellitus (T2DM). [1] J. Hagemeier, M. R. Woodward, U. A. Rafique et al., “Odor Functional enrichment analysis revealed that DEGs with identification deficit in mild cognitive impairment and upregulation primarily took part in and were enriched in the Alzheimer’s disease is associated with hippocampal and deep immune response, neutrophil degranulation, lysosome, oste- gray matter atrophy,” Psychiatry Research, vol. 255, pp. 87– ff ’ oclast di erentiation, and so on. Mishra and Brinton s 93, 2016. fl research has pointed out that the in ammatory immune [2] J. Shin, S. Y. Lee, S. H. Kim, Y. B. Kim, and S. J. Cho, “Multi- response is the unifying factor that connects each risk factor tracer PET imaging of amyloid plaques and neurofibrillary of AD [29]. The DEGs with downregulation exhibit a signif- tangles in Alzheimer’s disease,” NeuroImage, vol. 43, no. 2, icant enrichment in peptide biosynthetic process, translation, pp. 236–244, 2008. 8 Computational and Mathematical Methods in Medicine

[3] J. Zheng, Aberrant Proteasome Pathway in Alzheimer’s Disease of differentially expressed genes,” Frontiers in Neuroence, at EarlyStage Revealed by Whole-Transcriptome Study, Emory vol. 12, 2019. University, 2014. [20] G. K. Bhatti, A. P. Reddy, P. H. Reddy, and J. S. Bhatti, “Life- [4] R. A. Sweet, D. A. Bennett, N. R. Graff-Radford, R. Mayeux, style modifications and nutritional interventions in aging- and the National Institute on Aging Late-Onset Alzheimer’s associated cognitive decline and Alzheimer’s disease,” Fron- Disease Family Study Group, “Assessment and familial aggre- tiers in Aging Neuroscience, vol. 11, 2020. ’ gation of psychosis in Alzheimer s disease from the National [21] L. Bica, P. J. Crouch, R. Cappai, and A. R. White, “Metallo- ’ Institute on Aging Late Onset Alzheimer s Disease Family complex activation of neuroprotective signalling pathways as ” – Study, Brain, vol. 133, no. 4, pp. 1155 1162, 2010. a therapeutic treatment for Alzheimer’s disease,” Molecular [5] Y. He, J. Wu, L. Zhou, Y. Chen, F. Li, and H. Qian, “Quantifi- BioSystems, vol. 5, no. 2, pp. 134–142, 2009. ’ cation of cognitive function in Alzheimer s disease based on [22] D. Bosco, A. Fava, M. Plastino, T. Montalcini, and A. Pujia, ” deep learning, Frontiers in Neuroscience, vol. 15, 2021. “Possible implications of insulin resistance and metab- [6] A. F. Miranda et al., “Alzheimer’s disease: characteristics and olism in Alzheimer’s disease pathogenesis,” Journal of Cellular guidelines in dentistry,” RGO. Revista Gaúcha de Odontologia & Molecular Medicine, vol. 15, no. 9, pp. 1807–1821, 2011. (Online), vol. 58, no. 1, 2010. [23] T. Pietras, “ peptide toxicity and oxidative stress [7] T. Rapp, A. Grand, C. Cantet et al., “Public financial support in the pathogenesis of Alzheimer’s disease,” Postepy Psychiatrii receipt and non-medical resource utilization in Alzheimer’s I Neurologii, vol. 16, no. 1, pp. 69–73, 2007. ” disease results from the PLASA study, Social Science & Med- [24] J. Xia, D. M. Rocke, G. Perry, and M. Ray, “Differential net- – icine, vol. 72, no. 8, pp. 1310 1316, 2011. work analyses of Alzheimer’s disease identify early events in [8] H. C. Hendrie, K. S. Hall, A. Ogunniyi, and S. Gao, “Alzhei- Alzheimer’s disease pathology,” International Journal of Alz- mer’s disease, genes, and environment: the value of interna- heimer’s Disease, vol. 2014, article 721453, pp. 1–18, 2014. ” tional studies, Canadian Journal of Psychiatry Revue [25] J. K. Lee and N. J. Kim, “Recent advances in the inhibition of – Canadienne De Psychiatrie, vol. 49, no. 2, pp. 92 99, 2004. p38 MAPK as a potential strategy for the treatment of Alzhei- [9] H. Sordet-Guépet and P. Manckoundia, “Alzheimer disease mer’s disease,” Molecules, vol. 22, no. 8, p. 1287, 2017. and environmental exposure to neurotoxic factors: a contro- [26] Y. Kajiwara, T. Schiff, G. Voloudakis et al., “A critical role for ” versy, British Journal of Medicine and Medical Research, human caspase-4 in endotoxin sensitivity,” The Journal of – vol. 4, no. 34, pp. 5298 5311, 2014. Immunology, vol. 193, no. 1, pp. 335–343, 2014. “ [10] Q. Zhang, T. Gao, Y. Luo et al., Transient focal cerebral ische- [27] S. Hwang, M. H. Disatnik, and D. Mochly-Rosen, “Impaired mia/reperfusion induces early and chronic axonal changes in GAPDH-induced mitophagy contributes to the pathology of ’ ” rats: its importance for the risk of Alzheimer s disease, PLoS Huntington's disease,” EMBO Molecular Medicine, vol. 7, One, vol. 7, no. 3, article e33722, 2012. no. 10, pp. 1307–1326, 2015. “ ” [11] H. S. Wang, Research progress of gene chip technology, [28] L. Mirabello, E. R. Macari, L. Jessop et al., “Whole-exome Journal of Anhui Agricultural Sciences, vol. 35, no. 8, article sequencing and functional studies identify RPS29 as a novel 2241, 2007. gene mutated in multicase Diamond-Blackfan anemia fami- [12] H. Zhou and Y. Wang, “Gene chip technology and its use in lies,” Blood, vol. 124, no. 1, pp. 24–32, 2014. ” the research on drug resistance mechanism of osteosarcoma, [29] A. Mishra and R. D. Brinton, “Inflammation: bridging age, Chinese Journal of Clinical Oncology, vol. 34, 2011. menopause and APOEε4 genotype to Alzheimer’s disease,” [13] X. R. Qiao, X. Zhang, L. Mu, J. Tian, and Y. du, “GRB2-associ- Frontiers in Aging Neuroscience, vol. 10, p. 312, 2018. ated binding protein 2 regulates multiple pathways associated [30] S. J. Beck, L. Guo, A. Phensy et al., “Deregulation of mitochon- ” with the development of prostate cancer, Oncology Letters, drial F1FO-ATP synthase via OSCP in Alzheimer’s disease,” vol. 20, no. 4, p. 99, 2020. Nature Communications, vol. 7, no. 1, article 11483, 2016. [14] L. Cao, S. Y. Hu, J. Pan et al., “Expression and clinical signifi- cance of long non-coding RNA AC002454.1 in children with acute leukemia,” Zhongguo Shi Yan Xue Ye Xue Za Zhi, vol. 28, no. 5, pp. 1433–1439, 2020. [15] Z. Wang, J. Zhou, N. Dong, and W. Li, “Diagnostic signifi- cance of miR-937 in peripheral blood mononuclear cells of Kawasaki disease,” Clinical Laboratory, vol. 65, no. 11, 2019. [16] S. T. Sherry, M. H. Ward, M. Kholodov et al., “dbSNP: the NCBI database of genetic variation,” Nucleic Acids Research, vol. 29, no. 1, pp. 308–311, 2001. [17] G. Dennis Jr., B. T. Sherman, D. A. Hosack et al., “DAVID: database for annotation, visualization, and integrated discov- ery,” Genome Biology, vol. 4, article R60, 2003. [18] C. von Mering, M. Huynen, D. Jaeggi, S. Schmidt, P. Bork, and B. Snel, “STRING: a database of predicted functional associa- tions between proteins,” Nucleic Acids Research, vol. 31, no. 1, pp. 258–261, 2003. [19] N. Wang, Y. Zhang, L. Xu, and S. Jin, “Relationship between Alzheimer’s disease and the immune system: a meta-analysis