Chapter 24
Identification of Potential MicroRNA Biomarkers
by Meta-analysis
Hongmei Zhu and Siu-wai Leung
Abstract
Meta-analysis statistically assesses the results (e.g., effect sizes) across independent studies that are con-
ducted in accordance with similar protocols and objectives. Current genomic meta-analysis studies do not
perform extensive re-analysis on raw data because full data access would not be commonplace, although the
best practice of open research for sharing well-formed data have been actively advocated. This chapter
describes a simple and easy-to-follow method for conducting meta-analysis of multiple studies without
using raw data. Examples for meta-analysis of microRNAs (miRNAs) are provided to illustrate the method.
MiRNAs are potential biomarkers for early diagnosis and epigenetic monitoring of diseases. A number of
miRNAs have been identified to be differentially expressed, i.e., overexpressed or underexpressed, under
diseased states but only a small fraction would be highly effective biomarkers or therapeutic targets of
diseases. The meta-analysis method as described in this chapter aims to identify the miRNAs that are
consistently found dysregulated across independent studies as biomarkers.
Key wordsmicroRNA, Noncoding RNAs, Meta-analysis, Quality assessment, Biomarkers, Differen-
tial expression, Early diagnosis1 Introduction
Meta-analysis is a statistical method for systematically synthesizing
the scientific reports and quantifying an overall results such as effect
size from multiple independent studies by weighing each one by its
reliability and credibility [1, 2]. Meta-analysis has been widely used
in synthesizing the evidence for efficacy of medical treatments and
performance of diagnostic tests. It is increasingly used in many
other fields of research, such as education, psychology, criminology,
ecology, and molecular biology [3, 4]. Depending on the types of
the included studies and the information provided therein, their
outcome measures should be comparable for a meta-analysis.Mohini Gore and Umesh B. Jagtap (eds.),Computational Drug Discovery and Design, Methods in Molecular Biology, vol. 1762,
https://doi.org/10.1007/978-1-4939-7756-7_24,©Springer Science+Business Media, LLC, part of Springer Nature 2018
Electronic supplementary material:The online version of this chapter (https://doi.org/10.1007/978-1-4939-
7756-7_24) contains supplementary material, which is available to authorized users.
473