Computer Science and Engineering, Department of

 

First Advisor

Juan Cui

Date of this Version

Spring 5-22-2021

Document Type

Article

Citation

Roland, Madadjim, "Using an integrative machine learning approach to study microRNA regulation networks in pancreatic cancer progression" (2021).

Comments

A THESIS Presented to the Faculty of The Graduate College at the University of Nebraska In Partial Fulfillment of Requirements For the Degree of Master of Science, Major: Computer Science, Under the Supervision of Professor Juan Cui. Lincoln, Nebraska: May, 2021

Copyright © 2021 Roland Madadjim

Abstract

With advances in genomic discovery tools, recent biomedical research has produced a massive amount of genomic data on post-transcriptional regulations related to various transcript factors, microRNAs, lncRNAs, epigenetic modifications, and genetic variations. In this direction, the field of gene regulation network inference is created and aims to understand the interactome regulations between these molecules (e.g., gene-gene, miRNA-gene) that take place to build models able to capture behavioral changes in biological systems. A question of interest arises in integrating such molecules to build a network while treating each specie in its uniqueness. Given the dynamic changes of interactome in chaotic systems (e.g., cancers) and the dramatic growth of heterogeneous data on this topic, building scalable models is crucial. Indeed, recovering a model that can capture the relationships within this data constitutes a major challenge. This thesis addresses this challenge by using an integrative network learning model based on gene expression data to elucidate miRNA – gene interactions in cancer progression. First, we present a pre-processing pipeline for miRNA-gene interactions based on De Novo approach. Second, we introduce a machine learning approach for data integration of multiple data types such microarray, RNA-seq and CLIP based miRNA-RNA interactions along with graphical model fusion. Last, we show how the latter enabled transforming static interactions into semi-conditional ones. In a case study of human pancreatic cancer, we have identified gene regulatory networks distinctly associated with four progressive stages with a list of 12 miRNA-gene conditional interactions; The functional analysis with focus on microRNA-mediated dysregulation revealed significant changes in major cancer hallmarks. The identified novel pathological signaling and metabolic processes shed light on the regulatory roles that microRNAs play in pancreatic cancer progression. We believe this integrative model can be a robust and effective discovery tool to further the understanding of key regulatory characteristics in complex biological systems.

Adviser: Juan Cui

Share

COinS