Identification of More Feasible MicroRNA-mRNA Interactions within Multiple Cancers Using Principal Component Analysis Based Unsupervised Feature Extraction.

Y-H Taguchi

doi:10.3390/ijms17050696

Abstract

MicroRNA(miRNA)–mRNA interactions are important for understanding many biological processes, including development, differentiation and disease progression, but their identification is highly context-dependent. When computationally derived from sequence information alone, the identification should be verified by integrated analyses of mRNA and miRNA expression. The drawback of this strategy is the vast number of identified interactions, which prevents an experimental or detailed investigation of each pair. In this paper, we overcome this difficulty by the recently proposed principal component analysis (PCA)-based unsupervised feature extraction (FE), which reduces the number of identified miRNA–mRNA interactions that properly discriminate between patients and healthy controls without losing biological feasibility. The approach is applied to six cancers: hepatocellular carcinoma, non-small cell lung cancer, esophageal squamous cell carcinoma, prostate cancer, colorectal/colon cancer and breast cancer. In PCA-based unsupervised FE, the significance does not depend on the number of samples (as in the standard case) but on the number of features, which approximates the number of miRNAs/mRNAs. To our knowledge, we have newly identified miRNA–mRNA interactions in multiple cancers based on a single common (universal) criterion. Moreover, the number of identified interactions was sufficiently small to be sequentially curated by literature searches.

Highlights

MicroRNA(miRNA) is short non-coding RNA with an approximate length of 22 nt
Within the Hepatocellular Carcinoma (HCC) dataset, between 269 messenger RNAs (mRNAs) probes and 58 miRNA probes identified as outliers, we have successfully reduced the number of identified miRNA–mRNA pairs (21 pairs, see Tables S1 and S2)
Before discussing the identified miRNA–mRNA pairs, we demonstrate the feasibility of miRNA/mRNA identification by principal component analysis (PCA)-based unsupervised feature extraction (FE)

Summary

Introduction

MicroRNA(miRNA) is short non-coding RNA with an approximate length of 22 nt. Its canonical function is to target specific messenger RNAs (mRNAs) and post-transcriptionally suppress their expression. miRNAs bind to the three prime untranslated region of target mRNAs and promote their degradation or interrupt their translation [1]. MiRNA–mRNA interactions can be identified by computational methods, these are generally sequence-based and cannot accommodate the context-dependency of miRNA–mRNA bindings. Given the context dependent nature of miRNA–mRNA binding, the computational identification ability of miRNA–mRNA interactions can be greatly improved by accounting for the gene expression/miRNA expression. Given that typical values of N and M comprise a few tens of thousands and a few thousands, respectively, the number of possible pairs reaches several million. This suggests that miRNA and mRNA expressions are extremely well correlated, with p-values as small as 10−9, and are undetectable in noisy biological datasets. These difficulties could be reduced by reducing the number of pairs in the investigation

Objectives

Methods

Results