A comparison of normalization techniques for microRNA microarray data

Normalization of expression levels applied to microarray data can help in reducing measurement error. Different methods, including cyclic loess, quantile normalization and median or mean normalization, have been utilized to normalize microarray data. Although there is considerable literature regarding normalization techniques for mRNA microarray data, there are no publications comparing normalization techniques for microRNA (miRNA) microarray data, which are subject to similar sources of measurement error. In this paper, we compare the performance of cyclic loess, quantile normalization, median normalization and no normalization for a single-color microRNA microarray dataset. We show that the quantile normalization method works best in reducing differences in miRNA expression values for replicate tissue samples. By showing that the total mean squared error are lowest across almost all 36 investigated tissue samples, we are assured that the bias correction provided by quantile normalization is not outweighed by additional error variance that can arise from a more complex normalization method. Furthermore, we show that quantile normalization does not achieve these results by compression of scale.     

Quality Assessment and Data Analysis for microRNA Expression Arrays

MicroRNAs are small (~22 nt) RNAs that regulate gene expression and play important roles in both normal and disease physiology. The use of microarrays for global characterization of microRNA expression is becoming increasingly popular and has the potential to be a widely used and valuable research tool. However, microarray profiling of microRNA expression raises a number of data analytic challenges that must be addressed in order to obtain reliable results. We introduce here a universal reference microRNA reagent set as well as a series of nonhuman spiked-in synthetic microRNA controls, and demonstrate their use for quality control and between-array normalization of microRNA expression data. We also introduce diagnostic plots designed to assess and compare various normalization methods. We anticipate that the reagents and analytic approach presented here will be useful for improving the reliability of microRNA microarray experiments.     

Evaluation of normalization methods in mammalian microRNA-Seq data

Simple total tag count normalization is inadequate for microRNA sequencing data generated from the next generation sequencing technology. However, so far systematic evaluation of normalization methods on microRNA sequencing data is lacking. We comprehensively evaluate seven commonly used normalization methods including global normalization, Lowess normalization, Trimmed Mean Method (TMM), quantile normalization, scaling normalization, variance stabilization, and invariant method. We assess these methods on two individual experimental data sets with the empirical statistical metrics of mean square error (MSE) and Kolmogorov-Smirnov (K-S) statistic. Additionally, we evaluate the methods with results from quantitative PCR validation. Our results consistently show that Lowess normalization and quantile normalization perform the best, whereas TMM, a method applied to the RNA-Sequencing normalization, performs the worst. The poor performance of TMM normalization is further evidenced by abnormal results from the test of differential expression (DE) of microRNA-Seq data. Comparing with the models used for DE, the choice of normalization method is the primary factor that affects the results of DE. In summary, Lowess normalization and quantile normalization are recommended for normalizing microRNA-Seq data, whereas the TMM method should be used with caution.     

Intra-Platform Repeatability and Inter-Platform Comparability of MicroRNA Microarray Technology

Over the last decade, DNA microarray technology has provided a great contribution to the life sciences. The MicroArray Quality Control (MAQC) project demonstrated the way to analyze the expression microarray. Recently, microarray technology has been utilized to analyze a comprehensive microRNA expression profiling. Currently, several platforms of microRNA microarray chips are commercially available. Thus, we compared repeatability and comparability of five different microRNA microarray platforms (Agilent, Ambion, Exiqon, Invitrogen and Toray) using 309 microRNAs probes, and the Taqman microRNA system using 142 microRNA probes. This study demonstrated that microRNA microarray has high intra-platform repeatability and comparability to quantitative RT-PCR of microRNA. Among the five platforms, Agilent and Toray array showed relatively better performances than the others. However, the current lineup of commercially available microRNA microarray systems fails to show good inter-platform concordance, probably because of lack of an adequate normalization method and severe divergence in stringency of detection call criteria between different platforms. This study provided the basic information about the performance and the problems specific to the current microRNA microarray systems.     

Identification of Suitable Reference Genes for qRT-PCR Analysis of Circulating microRNAs in Hepatitis B Virus-Infected Patients

Circulating microRNAs (miRNAs) were found to exist in serum/plasma in a highly stable, cell-free form, and aberrantly expressed in many human diseases. Currently, the expression levels of circulating miRNAs are estimated by quantitative real-time polymerase chain reaction. However, no study has systematically evaluated reference genes for evaluating circulating microRNA expression. This study describes the identification and characterization of an appropriate reference gene for the normalization of circulating miRNA levels in hepatitis B virus (HBV)-infected patients and healthy people. Ten miRNAs that resemble the mean expression of the TaqMan low density array together with U6, RNU6B, and miR-16 were validated with two algorithms, geNorm, and NormFinder, after ensuring their equivalent expression between the two study groups. The combination of miR-26a, miR-221, and miR-22* is recommended as the most stable set of reference genes for circulating miRNA evaluation in HBV patients and healthy people.     

Normalization of microRNA expression levels in quantitative RT-PCR assays: identification of suitable reference RNA targets in normal and cancerous human solid tissues

Proper normalization is a critical but often an underappreciated aspect of quantitative gene expression analysis. This study describes the identification and characterization of appropriate reference RNA targets for the normalization of microRNA (miRNA) quantitative RT-PCR data. miRNA microarray data from dozens of normal and disease human tissues revealed ubiquitous and stably expressed normalization candidates for evaluation by qRT-PCR. miR-191 and miR-103, among others, were found to be highly consistent in their expression across 13 normal tissues and five pair of distinct tumor/normal adjacent tissues. These miRNAs were statistically superior to the most commonly used reference RNAs used in miRNA qRT-PCR experiments, such as 5S rRNA, U6 snRNA, or total RNA. The most stable normalizers were also highly conserved across flash-frozen and formalin-fixed paraffin-embedded lung cancer tumor/NAT sample sets, resulting in the confirmation of one well-documented oncomir (let-7a), as well as the identification of novel oncomirs. These findings constitute the first report describing the rigorous normalization of miRNA qRT-PCR data and have important implications for proper experimental design and accurate data interpretation.     

miR-Sens--a retroviral dual-luciferase reporter to detect microRNA activity in primary cells

MicroRNA-mRNA interactions are commonly validated and deconstructed in cell lines transfected with luciferase reporters. However, due to cell type-specific variations in microRNA or RNA-binding protein abundance, such assays may not reliably reflect microRNA activity in other cell types that are less easily transfected. In order to measure miRNA activity in primary cells, we constructed miR-Sens, a MSCV-based retroviral vector that encodes both a Renilla luciferase reporter gene controlled by microRNA binding sites in its 3' UTR and a Firefly luciferase normalization gene. miR-Sens sensors can be efficiently transduced in primary cells such as human fibroblasts and mammary epithelial cells, and allow the detection of overexpressed and, more importantly, endogenous microRNAs. Notably, we find that the relative luciferase activity is correlated to the miRNA expression, allowing quantitative measurement of microRNA activity. We have subsequently validated the miR-Sens 3' UTR vectors with known human miRNA-372, miRNA-373, and miRNA-31 targets (LATS2 and TXNIP). Overall, we observe that miR-Sens-based assays are highly reproducible, allowing detection of the independent contribution of multiple microRNAs to 3' UTR-mediated translational control of LATS2. In conclusion, miR-Sens is a new tool for the efficient study of microRNA activity in primary cells or panels of cell lines. This vector will not only be useful for studies on microRNA biology, but also more broadly on other factors influencing the translation of mRNAs.     

Systematic evaluation of three microRNA profiling platforms: microarray, beads array, and quantitative real-time PCR array

Background

A number of gene-profiling methodologies have been applied to microRNA research. The diversity of the platforms and analytical methods makes the comparison and integration of cross-platform microRNA profiling data challenging. In this study, we systematically analyze three representative microRNA profiling platforms: Locked Nucleic Acid (LNA) microarray, beads array, and TaqMan quantitative real-time PCR Low Density Array (TLDA).

Methodology/Principal Findings

The microRNA profiles of 40 human osteosarcoma xenograft samples were generated by LNA array, beads array, and TLDA. Results show that each of the three platforms perform similarly regarding intra-platform reproducibility or reproducibility of data within one platform while LNA array and TLDA had the best inter-platform reproducibility or reproducibility of data across platforms. The endogenous controls/probes contained in each platform have been observed for their stability under different treatments/environments; those included in TLDA have the best performance with minimal coefficients of variation. Importantly, we identify that the proper selection of normalization methods is critical for improving the inter-platform reproducibility, which is evidenced by the application of two non-linear normalization methods (loess and quantile) that substantially elevated the sensitivity and specificity of the statistical data assessment.

Conclusions

Each platform is relatively stable in terms of its own microRNA profiling intra-reproducibility; however, the inter-platform reproducibility among different platforms is low. More microRNA specific normalization methods are in demand for cross-platform microRNA microarray data integration and comparison, which will improve the reproducibility and consistency between platforms.     

Assessment of Reference Genes for Real-Time Quantitative PCR Gene Expression Normalization During C2C12 and H9c2 Skeletal Muscle Differentiation

Skeletal muscle differentiation occurs during muscle development and regeneration. To initiate and maintain the differentiated state, a multitude of gene expression changes occur. Accurate assessment of these differentiation-related gene expression changes requires good quality template, but more specifically, appropriate internal controls for normalization. Two cell line-based models used for in vitro analyses of muscle differentiation incorporate mouse C2C12 and rat H9c2 cells. In this study, we set out to identify the most appropriate controls for mRNA expression normalization during C2C12 and H9c2 differentiation. We assessed the expression profiles of Actb, Gapdh, Hprt, Rps12 and Tbp during C2C12 differentiation and of Gapdh and Rps12 during H9c2 differentiation. Using NormFinder, we validated the stability of the genes individually and of the geometric mean generated from different gene combinations. We verified our results using Myogenin. Our study demonstrates that using the geometric mean of a combination of specific reference genes for normalization provides a platform for more precise test gene expression assessment during myoblast differentiation than using the absolute expression value of an individual gene and reinforces the necessity of reference gene validation.     

基于TransformerMGI的microRNA靶向基因预测

生物小分子microRNA可以对基因表达进行正向或负向调控,研究microRNA与基因之间的关系对于机体稳态的维持和疾病治疗都有着重要意义。利用深度学习方法对microRNA和基因靶向关系进行预测,提出了TransformerMGI模型。在特征工程阶段,针对生物序列潜在信息难以准确地提取这一问题,TransformerMGI模型分别采用了基于图卷积神经网络的GP-GCN方法和DNA2Vec模型对microRNA和基因数据的潜在信息进行提取,得到了二者的表征嵌入矩阵,在模型方面,TransformerMGI模型引入了幂归一化来改进经典的深度学习模型。利用microRNA和基因数据经过特征提取后得到两个表征矩阵,这两个矩阵分别被放入TransformerMGI模型中,通过TransformerMGI模型内部的Attention机制对二者自身和相互的特征信息进行了聚合和关联运算,最终预测出microRNA调控基因的概率。采用ROC曲线下面积和准确召回率曲线作为模型性能评价指标,将TransformerMGI与其他现有模型进行了比较评估。实验结果表明,TransformerMGI模型的AUC和AUPRC评分均可达0.91以上,优于现有的其他模型。TransformerMGI模型能在不考虑生物学原理和基因组背景的前提下,仅依赖microRNA和基因的碱基序列信息,实现microRNA靶向基因的预测,从而为后续的microRNA靶向基因预测研究提供了可借鉴的深度学习方法。     

RNAi and microRNA: breakthrough technologies for the improvement of plant nutritional value and metabolic engineering

This article raises the complex issue of improving plant nutritional value through metabolic engineering and the potential of using RNAi and micro RNA technologies to overcome this complexity, focusing on a few key examples. It also highlights current knowledge of RNAi and microRNA functions and discusses recent progress in the development of new RNAi vectors and their applications. RNA interference (RNAi) and microRNA (miRNA) are recent breakthrough discoveries in the life sciences recognized by the 2006 Nobel Prize in Physiology or Medicine. The importance of these discoveries relates not only to elucidating the fundamental regulatory aspects of gene expression, but also to the tremendous potential of their applications in plants and animals. Here, we review recent applications of RNAi and microRNA for improving the nutritional value of plants, discuss applications of metabolomics technologies in genetic engineering, and provide an update on the related RNAi and microRNA technologies.     

Contrast normalization of oligonucleotide arrays.

Affymetrix high-density oligonucleotide array is a tool that has the capacity to simultaneously measure the abundance of thousands of mRNA sequences in biological samples. In order to allow direct array-to-array comparisons, normalization is a necessity. When deciding on an appropriate normalization procedure there are a couple questions that need to be addressed, e.g., on which level should the normalization be performed: On the level of feature intensities or on the level of expression indexes? Should all features/expression indexes be used or can we choose a subset of features likely to be unregulated? Another question is how to actually perform the normalization: normalize using the overall mean intensity or use a smooth normalization curve? Most of the currently used normalization methods are linear; e.g., the normalization method implemented in the Affymetrix software GeneChip is based on the overall mean intensity. However, along with alternative methods of summarizing feature intensities into an expression index, nonlinear methods have recently started to appear. For many of these alternative methods, the natural choice is to normalize on the level of feature intensities, either using all feature intensities or only perfect match intensities. In this report, a nonlinear normalization procedure aimed for normalizing feature intensities is proposed.     

Suitable internal control microRNA genes for measuring miRNA abundance in pig milk during different lactation periods

Determination of an optimal set/number of internal control microRNA (miRNA) genes is a critical, but often undervalued, detail of quantitative gene expression analysis. No validated internal genes for miRNA quantitative PCR (q-PCR) in pig milk were available. We compared the expression stability of six porcine internal control miRNA genes in pig milk from different lactation periods (1 h, 3 days, 7 days, 14 days, 21 days, and 28 days postpartum), using an EvaGreen q-PCR approach. We found that using the three most stable internal control genes to calculate the normalization factor is sufficient for producing reliable q-PCR expression data. We also found that miRNAs are superior to ribosomal RNA (rRNA) and snRNA, which are commonly used as internal controls for normalizing miRNA q-PCR data. In terms of economic and experimental feasibility, we recommend the use of the three most stable internal control miRNA genes (miR-17, -107 and -103) for calculating the normalization factors for pig milk samples from different lactation periods. These results can be applied to future studies aimed at measuring miRNA abundance in porcine milk.