SLIIT International Conference on Advancements in Sciences and Humanities [SICASH] 2023

Permanent URI for this collectionhttps://rda.sliit.lk/handle/123456789/3589

Browse

Filter results by typing the first few letters
Now showing 1 - 1 of 1
  • Thumbnail Image
    PublicationOpen Access
    Determining Differentially Expressed Genes in Dengue Patients during Disease Progression
    (Faculty of Humanities and Sciences, SLIIT, 2023-11-01) Coorey, H.; Jayatillake, R.; Jayathilaka, N.; Ambanpola, N.
    Gene expression studies on gene transcription to synthesize functional gene products have been used extensively to understand the biological differences between different disease conditions. Thus, this study determines differentially expressed genes in dengue infection during disease progression following the three phases: Febrile, Defervescence and Convalescent. Integrative data analysis of two publicly available longitudinal datasets in the Gene Expression Omnibus (GEO) database has been employed to accomplish the prime objective of exploring temporal gene expression patterns. The Friedman test was given more emphasis due to the non-normality distributions of data. Since previous studies on gene expression have not primarily relied on normality assumption, repeated measures analysis of variance and linear mixed models were implemented to examine the potential of detecting differentially expressed genes despite non-normality. The Friedman test indicated that gene expression levels differentiate with different phases in dengue disease over time, resulting in a high number of significant differentially expressed genes compared to the other two techniques. The pathway analysis approach consists of significant differentially expressed genes derived from the Friedman test. The results identified 27 and 26 upregulated pathways for the “Febrile and Convalescent” and “Defervescence and Convalescent” groups respectively. Moreover, genes available in pathways were not identified by the two parametric tests for non-normal data implying that the parametric approaches resulted in the least significance for data with non-normal distributions.