Skip to main page content
U.S. flag

An official website of the United States government

Dot gov

The .gov means it’s official.
Federal government websites often end in .gov or .mil. Before sharing sensitive information, make sure you’re on a federal government site.

Https

The site is secure.
The https:// ensures that you are connecting to the official website and that any information you provide is encrypted and transmitted securely.

Access keys NCBI Homepage MyNCBI Homepage Main Content Main Navigation
. 2015 Oct 12:16:767.
doi: 10.1186/s12864-015-1971-9.

Integrated ordination of miRNA and mRNA expression profiles

Affiliations

Integrated ordination of miRNA and mRNA expression profiles

Giacomo Diaz et al. BMC Genomics. .

Abstract

Background: Several studies have investigated miRNA and mRNA co-expression to identify regulatory networks at the transcriptional level. A typical finding of these studies is the presence of both negative and positive miRNA-mRNA correlations. Negative correlations are consistent with the expected, faster degradation of target mRNAs, whereas positive correlations denote the existence of feed-forward regulations mediated by transcription factors. Both mechanisms have been characterized at the molecular level, although comprehensive methods to represent miRNA-mRNA correlations are lacking. At present, genome-wide studies are able to assess the expression of more than 1000 mature miRNAs and more than 35,000 well-characterized human genes. Even if studies are generally restricted to a small subset of genes differentially expressed in specific diseases or experimental conditions, the number of potential correlations remains very high, and needs robust multivariate methods to be conveniently summarized by a small set of data.

Results: Nonparametric Kendall correlations were calculated between miRNAs and mRNAs differentially expressed in livers of patients with acute liver failure (ALF) using normal livers as controls. Spurious correlations due to the histopathological composition of samples were removed by partial correlations. Correlations were then transformed into distances and processed by multidimensional scaling (MDS) to map the miRNA and mRNA relationships. These showed: (a) a prominent displacement of miRNA and mRNA clusters in ALF livers, as compared to control livers, indicative of gene expression dysregulation; (b) a clustering of mRNAs consistent with their functional annotations [CYP450, transcription factors, complement, proliferation, HLA class II, monocytes/macrophages, T cells, T-NK cells and B cells], as well as a clustering of miRNAs with the same seed sequence; and (c) a tendency of miRNAs and mRNAs to populate distinct regions of the MDS plot. MDS also allowed to visualize the network of miRNA-mRNA target pairs.

Conclusions: Different features of miRNA and mRNA relationships can be represented as thematic maps within the framework of MDS obtained from pairwise correlations. The symmetric distribution of positive and negative correlations between miRNA and mRNA expression suggests that miRNAs are involved in a complex bidirectional molecular network, including, but not limited to, the inhibitory regulation of miRNA targets.

PubMed Disclaimer

Figures

Fig. 1
Fig. 1
Frequency distribution comprehensive of all pairwise miRNA-mRNA, miRNA-miRNA and mRNA-mRNA Kendall correlations. a ALF livers. b normal livers. c ALF livers after partial correlations calculated for the level of necrosis
Fig. 2
Fig. 2
Correlation between gene expression and level of necrosis in ALF livers. The plots show a selection of representative mRNAs positively (left) or negatively (right) correlated with the degree of necrosis in ALF livers with |R| > 0.9. The level of hepatic necrosis is on the X axis; the gene expression is on the Y axis. Presumably, mRNAs positively correlated with necrosis are produced by non-hepatocyte cells, whereas those negatively correlated with necrosis are produced by hepatocytes. Gene expressions were standardized to fit the same scale range. Multiple dots of the same gene (color) for each level of necrosis represent data of multiple samples
Fig. 3
Fig. 3
Gene expression corrected for necrosis. Data refer to the same mRNAs shown in Fig. 2. a control livers. b ALF livers. c ALF livers data expected for zero necrosis, obtained as the intercept of the regression between necrosis (X variable) and gene expression (Y variable). d correlation between control livers and original ALF livers data. e correlation between control livers and ALF livers data expected for zero necrosis
Fig. 4
Fig. 4
MDS mapping of mRNAs differentially expressed in ALF livers. MDS was applied to a 640 × 640 square matrix including the Kendall correlations among 109 miRNAs and 531 mRNAs differentially expressed in ALF livers. This figure shows only a subset of 87 mRNAs functionally related to CYP450, transcription factors, complement, HLA class II, monocytes/macrophages, T cells, T-NK cells and B cells. These nine functional classes are delimited by dispersion ellipses with a confidence of 1.6 standard deviations. A 360° rotation of 3D ellipses is shown in the Additional file 2: Movie 1. The nine ellipses are also shown in the next Figs. 5, 6, 7 and 8 for reference. The numbers in parentheses, on the right of gene symbols, are the fold changes of original data, not corrected for necrosis. A prominent segregation of leukocyte-related mRNAs from hepatocyte-related mRNAs is apparent
Fig. 5
Fig. 5
MDS density plot of mRNAs differentially expressed in ALF livers. The density plot was calculated for the MDS map of 531 mRNAs differentially expressed in ALF livers, including the 87 mRNAs of Fig. 3 (shown by large dots, symbols, original fold changes and dispersion ellipses) and the remaining 444 mRNAs (shown by small dots)
Fig. 6
Fig. 6
MDS density plot of miRNAs differentially expressed in ALF livers. The numbers in parentheses, on the right of miRNA symbols, are the fold changes of original data, not corrected for necrosis. The 17 green-outlined points are the miRNAs whose median correlation with mRNAs decreased more than 0.15 Kendall tau in ALF livers, whereas the red-outlined point is the only miRNA whose median correlation increased more than 0.15 Kendall tau (Additional file 4: Figure S1). The dispersion ellipses of functional mRNA clusters are shown for reference. The inset shows the complementarity of miRNA (cyan) and mRNA (yellow) MDS density plots
Fig. 7
Fig. 7
MDS mapping of the network of miRNAs and target mRNAs differentially expressed in ALF livers. For clarity, only the 87 mRNAs of the nine functional groups are shown. Target mRNAs were obtained from the microRNA.org database, selecting miRNA-mRNA pairs with conserved miRNAs and a good (<= −0.1) mirSVR score. The dispersion ellipses of functional mRNA clusters are shown for reference
Fig. 8
Fig. 8
MDS mapping of miRNAs with the same seed sequence, found among miRNAs differentially expressed in ALF livers. MiRNAs with the same seed sequence are encircled by small ellipses. Small blue and red ellipses indicate down-regulated and up-regulated miRNAs, respectively. The dispersion ellipses of functional mRNA clusters are also shown for reference. The complete sequence of these miRNAs is shown in the Additional file 5: Table S3
Fig. 9
Fig. 9
MDS mapping of mRNAs expressed in control livers. This figure is to be compared with Fig. 4. A 360° rotation of 3D ellipses is shown in the Additional file 3: Movie 2. The nine ellipses are also shown in next Figs. 10, 11, 12 and 13 for reference. The numbers in parentheses, on the right of gene symbols, are the fold changes
Fig. 10
Fig. 10
MDS density plot of mRNAs expressed in control livers. This figure is to be compared with Fig. 5
Fig. 11
Fig. 11
MDS density plot of miRNAs expressed in control livers. This figure is to be compared with Fig. 6
Fig. 12
Fig. 12
MDS mapping of the network of miRNAs and target mRNAs expressed in control livers. This figure is to be compared with Fig. 7
Fig. 13
Fig. 13
MDS mapping of miRNAs with the same seed sequence expressed in control livers. This figure is to be compared with Fig. 8

Similar articles

Cited by

References

    1. Filipowicz W, Bhattacharyya SN, Sonenberg N. Mechanisms of post-transcriptional regulation by microRNAs: are the answers in sight? Nat Rev Genet. 2008;9(2):102–14. doi: 10.1038/nrg2290. - DOI - PubMed
    1. Valencia-Sanchez MA, Liu J, Hannon GJ, Parker R. Control of translation and mRNA degradation by miRNAs and siRNAs. Genes Dev. 2006;20(5):515–24. doi: 10.1101/gad.1399806. - DOI - PubMed
    1. Cascione L, Gasparini P, Lovat F, Carasi S, Pulvirenti A, Ferro A, Alder H, He G, Vecchione A, Croce CM, et al. Integrated microRNA and mRNA signatures associated with survival in triple negative breast cancer. PLoS One. 2013;8(2):e55910. doi: 10.1371/journal.pone.0055910. - DOI - PMC - PubMed
    1. Nunez-Iglesias J, Liu CC, Morgan TE, Finch CE, Zhou XJ. Joint genome-wide profiling of miRNA and mRNA expression in Alzheimer’s disease cortex reveals altered miRNA regulation. PLoS One. 2010;5(2):e8898. doi: 10.1371/journal.pone.0008898. - DOI - PMC - PubMed
    1. Van der Auwera I, Limame R, van Dam P, Vermeulen PB, Dirix LY, Van Laere SJ. Integrated miRNA and mRNA expression profiling of the inflammatory breast cancer subtype. Br J Cancer. 2010;103(4):532–41. doi: 10.1038/sj.bjc.6605787. - DOI - PMC - PubMed

Publication types