This is a wonderful question. Initially my ideas about the multidimensional surface of regulatory events was too simplistic and I assumed that regulation arises randomly as noise that is fine-tuned as feeble signals because of additive probabilities that are negligible in themselves. My opinion was based mainly on the fact that regulation is a complicated event depending on multitude of parameters that ought to follow a non-linear trajectory. for example, transcriptional regulation is a complex event involving DNA modifications ( as in CpG islands), protein regulation ( as in PTMs of histones that form the chromatin and allosteric regulation either by protein-protein interaction or by small-molecule protein interaction regulating the compaction process and accessibility of DNA for transcription) and each one of these processes regulating DNA winding and unwinding resulting in "transcriptional regulation" is a complex event of transcriptional regulation, translational regulation and protein regulation controlled both temporally and spatially as the half life of the nucleic acid and protein players. For a long time I thought that teasing out such complex network into individual components might not be feasible to assign a hierarchy to such regulation vis-a-vis its phenotype. Though I still hold that view, recent findings on the complexity of a multicellular organism and the amount of "junk" DNA in its genome being highly correlated with transcriptional regulation may be indicative of possible hierarchy of regulational surface ( Though I am still skeptical about the existence of such hierarchy). Further, a recent nature paper showing that primary transcripts of microRNAs encode regulatory peptides indicate towards how intertwined transcriptional regulation and translational regulation might be (Nature, 520, 90–93, 2015). All this recent findings may, in the future, shed some light on the nature of regulational landscape. However, as of now, I think that we are staring at 10 odd pieces of a jigsaw puzzle that ought to have 1000 pieces. Leave aside the exercise of assembling the jigsaw puzzle, which in itself might be a challenge, our goal should be to first search out the 990 missing pieces. I hope the answer makes sense. :)

In signalling pathways, the most downstream  genes are transcriptionally regulated, and the upstream proteins, by PTM, being the most upstream allosterically reg. 

Dear Carrasco, Thanks for the information. Can you send me any references wherein such a pattern is observed in any signalling pathways?

I agree with Bharath.  There is a lot more than genes regulating expression and influencing phenotoype.  The entire microenvironment regulates expression and thus phenotype.  That is why you can take embryonal carcinoma cells and inject them subcutaneously into mice, resulting in teratocarcinomas, but inject the same cells into a blastocyst and give rise to normal chimeric mice.  There are many experimental examples of cancer cells reverting to normal when injected into normal animals.   There are multiple factors regulating gene expression and many of them function through epigenetic mechanisms which are strongly influenced by the environment.  And yes, it is absolutely nonlinear.  So what regulates a particular biological phenotype is enormously complex and cannot be generalized from one phenotype to another.