Please use this identifier to cite or link to this item: http://hdl.handle.net/10397/84017
Title: Structural analysis of gene co-expression in chronic myelogenous leukemia
Authors: Wang, Fengfeng
Degree: Ph.D.
Issue Date: 2014
Abstract: Background: Chronic myelogenous leukemia (CML) is a clonal myeloproliferative disorder characterized by an increased proliferation of granulocytes in bone marrow. The characteristics of CML at the cellular level include increased proliferation, increased resistance to apoptosis and alterations in adhesion properties. Co-expression analysis has been used to study functionally related genes, since the co-expressed genes are more likely to participate in the similar biological processes and signal pathways. Moreover, researchers found that genes with similar mRNA expression profiles tend to be regulated via the same mechanism (s), e.g. the same regulator. We plan to explore the differences between the normal and the CML groups in the co-expression patterns of those genes involved in a functional gene set, regulated by the same regulators, and covering from the whole genome, in order to further explore the altered biological pathways and novel mechanisms in CML. Nucleophosmin 1 (NPM1) is important in ribosomal synthesis and malignancies. The NPM1-associated gene set was chosen as the candidate gene set for the co-expression analysis. We wonder if NPM1-associated genes can affect the ribosomal synthesis and translation process in CML. E2F13 and MYC are important transcription factors (TFs) reciprocally regulated in the transcription process to form positive feedback loops. Target genes regulated by E2F13 or MYC are related to cell proliferation and apoptosis. MicroRNAs (miRNAs) are post-transcriptional regulators regulating target gene expression. Mature miRNAs from the miR-17-92 cluster are overexpressed in chronic-phase CML patients compared with normal individuals. The overexpression can promote cell cycle progression and proliferation, and inhibit apoptosis. We wonder what the co-expression patterns of the target genes directly reglated by E2F13 and MYC, or by miRNAs in the normal and the CML groups are.
Result and conclusion: We presented a distribution-based approach for gene pair classification by identifying a disease-specific cutoff point that classified the co-expressed gene pairs into strong and weak classes. Our developed method effectively identified the differences in the co-expression patterns from the overall structure: a) whole genome co-expression analysis: p-value < 0.05 for the maximum deviation D = 0.041; b) NPM1-associated gene set co-expression analysis: p-value = 1.71x10⁻²² < 0.05 for the maximum deviation D = 0.109; c) E2F and MYC target genes co-expression analysis: p-value = 2.00x10⁻³⁴ < 0.05 for the maximum deviation D = 0.0577; d) miR-17-92 cluster target genes co-expression analysis: p-value = 2.62x10⁻⁵⁸ < 0.05 for the maximum deviation D = 0.0567. The distribution-based classification divided the co-expressed gene pairs into specific and common groups, forming the co-expression structures. Functional annotation showed that ribosomal protein (RP) genes were more likely to be co-expressed in the CML group compared to the normal group. In addition, genes involved in the ribosomal synthesis and translation process tended to be co-expressed in the CML group. While, genes related to cell adhesion and angiogenesis properties, as well as metabolism processes were more likely to be co-expressed in the normal group. The co-expression pattern in the normal group represents the healthy pathological balance. Our findings may be helpful in exploring the underlying mechanisms of CML, and provide useful information in cancer treatment.
Subjects: Chronic myeloid leukemia -- Genetic aspects.
Gene expression.
Hong Kong Polytechnic University -- Dissertations
Pages: xxiv, 174 pages : illustrations (chiefly color) ; 30 cm
Appears in Collections:Thesis

Show full item record

Page views

57
Last Week
0
Last month
Citations as of Mar 24, 2024

Google ScholarTM

Check


Items in DSpace are protected by copyright, with all rights reserved, unless otherwise indicated.