We further derived the level of free Ago\miRNA complexes at which a specific target is halfway between its maximum level data with the computational model (Fig?3A), added noise in target levels comparable to the noise observed (Fig?3B), and then experimented with the selection of cells to use in the inference and with the smoothing of the target levels (Fig?3B and C) to most accurately recover the input parameters (see also Materials and Methods and Appendix?Fig S4). expression of two distinct miRNAs was induced over a wide range, we have inferred parameters describing the response of LGB-321 HCl hundreds of miRNA targets to miRNA induction. Individual targets have widely different response dynamics, and only a small proportion of predicted targets exhibit high sensitivity to miRNA induction. Our data reveal for the first time the response parameters of the entire network of endogenous miRNA targets to miRNA induction, demonstrating that miRNAs correlate target expression and at the same time increase the variability in expression of individual targets across cells. The approach is generalizable to other miRNAs and post\transcriptional regulators to improve the understanding of gene expression dynamics in individual cell types. have striking developmental phenotypes (Ha most miRNA genes are individually dispensable for development and viability, at least in the worm (Miska measurements indicate that miRNA target sites can have widely different affinities for the miRNACArgonaute complex (Wee miRNACtarget interaction constants are lacking. Taking advantage of a system in which the expression of a single miRNA precursor can be induced over a wide concentration range, we measured the transcriptomes of thousands of individual cells and assessed how the expression levels of miRNA targets relate to the expression level of the miRNA. We obtained experimental evidence for behaviors that were previously suggested by computational models or evaluated only with miRNA target reporters. These include the non\linear, ultrasensitive response of miRNA targets to changes in the miRNA concentration as well as the dependency of?the variability in target levels between cells on the concentration LGB-321 HCl of the miRNA. Furthermore, we found that only a small fraction of predicted targets are highly sensitive to changes in miRNA expression. With a computational model, we illustrate how these targets can influence the expression of other targets as competing RNAs. Our approach is applicable to other post\transcriptional regulators of mRNA stabilityallowing the analysis of their concentration\dependent impact on the transcriptome. Results A system to study the impact of miRNA expression on the transcriptome of individual cells miRNA target reporters are widely used to study miRNA\dependent?gene regulation. However, these reporters are often expressed at much higher levels than when expressed from their corresponding genomic loci. Furthermore, these reporters do not respond to the regulatory influences to which the endogenous transcripts respond. To circumvent these issues and investigate the crosstalk of miRNA targets in their native expression context, we used a human embryonic kidney (HEK) 293 cell line, i199 (Hausser CIT of log2 expression values?=?0.89, (2013) predicted that LGB-321 HCl the coefficient of variation (CV) of miRNA targets increases with the transcription rate of the miRNA, showing a local maximum in the region where the miRNA and targets are in equimolar ratio. The correlation of expression levels of mRNAs that are targeted by the same miRNA was predicted to exhibit a maximum around the same threshold. We used a similar simple model of miRNA\dependent gene regulation to predict the behavior of targets in our experimental system. Briefly, we considered mRNA targets of a given LGB-321 HCl miRNA, each with a specific transcription rate could bind a miRNA\containing Argonaute (Ago) complex at rate and dissociate from the complex at rate of Ago\miRNA complexes in a given cell was constant, though varying between cells. The number of free Ago\miRNA complexes is then given by differential equations targets to miRNA induction is shown in Fig?2A. Figure?2B and C shows the variability of target expression between simulated cells and the pairwise correlations of target expression levels across all simulated cells, as functions of the total miRNA level. Similar to the predictions of Bosia (2013), the targets in our system also experience destabilization, increased correlation, and increased expression noise, all within a limited range of miRNA expression, i.e. at a specific threshold. Figure?2B also shows that for each target, the coefficient of variation increases in function of miRNA expression level, as the target.