Here are some examples of real gene-list datasets analyzed by miREM to discover the actual increase or decrease in miRNA(s) activity.
1. miRNA knock-in experiments (from Eichhorn et al. Mol. Cell, 2014)
As a proof of concept, we used miREM on gene-lists derived from knock-in miRNA experiments acquired from total/cytoplasmic & t/ribo depleted RNAseq in U2OS cell lines.
For all these experiments, miREM is able to predict the knock-in RNA with the highest EM score.
i. Downregulated genes derived from hsa-miR-155 knock-in experiment (total RNA) (gene-list | results) ii. Downregulated genes derived from hsa-miR-155 knock-in experiment (cytoplasmic RNA) (gene-list | results) iii. Downregulated genes derived from hsa-miR-155 knock-in experiment (t/ribo RNA depleted) (gene-list | results) _ iv. Downregulated genes derived from hsa-miR-1 knock-in experiment (total RNA) (gene-list | results) v. Downregulated genes derived from hsa-miR-1 knock-in experiment (cytoplasmic RNA) (gene-list | results) vi. Downregulated genes derived from hsa-miR-1 knock-in experiment (t/ribo RNA depleted) (gene-list | results)
2. miRNA double knockout experiment (from Yu et al. Genes & Dev., 2010)
miREM is also able to predict the co-repressive actions of several miRNAs, as shown by the following analysis using a dataset from Yu et al., where the bicistronic miR-144/451 was removed.
NB. miREM’s heatmap (see results) illustrates the co-repressive actions of miR-144/451.
i. Predicting co-regulated miRNAs in knockout cluster miR-144/451 (gene-list | results)
3. miRNA double knockout experiment (from Henao-Mejia et al. Immunity., 2013)
miREM is also able to predict miR-181a and miR-181b double knocked out in the experiment.
i. Predicting co-regulated miRNAs in knockout cluster miR-181a1/b1 (gene-list | results)
Footnotes
In case study 1, RNA-seq raw reads were downloaded and aligned to hg38 using STAR aligner with default parameters [1]. Gene expression levels and differentially expressed gene-list were determined using HOMER package [2] with p-value threshold setup at 0.05 (full list with fold changes and p-values are available from the miREM portal under the section “Documentation>Examples”). In case study 2, the up-regulated genes are derived from an analysis performed by GEO2R a NCBI toolkit (https://www.ncbi.nlm.nih.gov/geo/geo2r/?acc=GSE21041), using default settings. Up-regulated genes were determined using the following threshold: log(fold-change) = 1.5; adjusted p-value = 0.5 (Benjamini and Hochberg). The gene-list used in case study 3 was acquired from the processed dataset provided by the original publication available in GEO datasets (GSE41090).
References:
[1]: Dobin A, Davis CA, Schlesinger F, Drenkow J, Zaleski C, Jha S, Batut P, Chaisson M, Gingeras TR. STAR: ultrafast universal RNA-seq aligner. Bioinformatics. 2013 Jan 1;29(1):15-21.
[2]: Heinz S, Benner C, Spann N, Bertolino E et al. Simple Combinations of Lineage-Determining Transcription Factors Prime cis-Regulatory Elements Required for Macrophage and B Cell Identities. Mol Cell 2010 May 28;38(4):576-589.