Clustering, Pathway Enrichment, and Protein-Protein Interaction Analysis of Gene Expression in Neurodevelopmental Disorders

Abstract

Neuronal developmental disorder is a class of diseases in which there is impairment of the central nervous system and brain function. The brain in its developmental phase undergoes tremendous changes depending upon the stage and environmental factors. Neurodevelopmental disorders include abnormalities associated with cognitive, speech, reading, writing, linguistic, communication, and growth disorders with lifetime effects. Computational methods provide great potential for betterment of research and insight into the molecular mechanism of diseases. In this study, we have used four samples of microarray neuronal developmental data: control, RV (resveratrol), NGF (nerve growth factor), and RV + NGF. By using computational methods, we have identified genes that are expressed in the early stage of neuronal development and also involved in neuronal diseases. We have used MeV application to cluster the raw data using distance metric Pearson correlation coefficient. Finally, 60 genes were selected on the basis of coexpression analysis. Further pathway analysis was done using the Metascape tool, and the biological process was studied using gene ontology database. A total of 13 genes AKT1, BAD, BAX, BCL2, BDNF, CASP3, CASP8, CASP9, MYC, PIK3CD, MAPK1, MAPK10, and CYCS were identified that are common in all clusters. These genes are involved in neuronal developmental disorders and cancers like colorectal cancer, apoptosis, tuberculosis, amyotrophic lateral sclerosis (ALS), neuron death, and prostate cancer pathway. A protein-protein interaction study was done to identify proteins that belong to the same pathway. These genes can be used to design potential inhibitors against neurological disorders at the early stage of neuronal development. The microarray samples discussed in this publication are part of the data deposited in NCBI's Gene Expression Omnibus (Yadav et al., 2018) and are accessible through GEO Series (accession number GSE121261).

Identification of potential hub genes via bioinformatics analysis combined with experimental verification in colorectal cancer

“…To further understand the function and enrichment pathway of differential genes, we divided the different genes into three parts: the whole genes, upregulated genes, and downregulated genes. Metascape (http://metascape.org/) is a web‐based tool for gaining insight into the biological functions of these coexpressed genes 20,21 . The GO vocabulary is divided into three parts: biological process (BP), cell component (CC), and molecular function (MF).…”

“…Metascape (http://metascape.org/) is a web-based tool for gaining insight into the biological functions of these coexpressed genes. 20,21 The GO vocabulary is divided into three parts: biological process (BP), cell component (CC), and molecular function (MF). The KEGG pathway was used to identify statistically significant signaling pathways, and P < .05 was considered as a statistically significant difference.…”

GRHL2 Expression Functions in Breast Cancer Aggressiveness and Could Serve as Prognostic and Diagnostic Biomarker for Breast Cancer

“…The following databases were used: Oncomine ( www.oncomine.org ), 20 TIMER ( http://cistrome.org/TIMER/ ), 21 CCLE ( https://portals.broadinstitute.org/ccle/about ), 22 GEPIA ( http://gepia.cancer-pku.cn/ ), 23 The Human Protein Atlas (HPA https://www.proteinatlas.org/ ), 24 UALCAN ( http://ualcan.path.uab.edu/ ), 25 TCGA database ( https://www.cancer.gov/tcga ), Kaplan–Meier Plotter ( https://kmplot.com/analysis/ ), 26 PrognoScan ( http://dna00.bio.kyutech.ac.jp/PrognoScan/index.html ), 27 LinkedOmics ( http://www.linkedomics.org/ ), 28 Metascape ( http://metascape.org/ ), 29 cBioPortal ( https://www.cbioportal.org/ ), 30 DiseaseMeth version 2.0 ( http://biobigdata.hrbmu.edu.cn/diseasemeth/ ), 31 MEXPRESS ( https://mexpress.be ), 32 and MethSurv ( https://biit.cs.ut.ee/methsurv/ ) 33 (see Supplementary Materials for instructions on using the databases).…”

Common cancer biomarkers of breast and ovarian types identified through artificial intelligence

“…Our analysis has combined gene expression clustering, network centrality, and pathway analysis to discover the most influential proteins among the overlapping breast and ovarian cancer biomarkers. Though some of these techniques have been recently used in combination to identify genes and pathways in hepatocellular carcinoma (Liu et al, 2016), depression (Le et al, 2018), and neurodevelopmental disorders (Yadav & Srivastava, 2018), this article marks the first usage of all three in consolidation. We have utilized gap statistics with k‐means and hierarchical clustering in our pipeline and both methods separately showed breast and ovarian cancer in the same cluster, enabling us to continue with further downstream analysis.…”