I think one should define what you mean by "regulatory genes". Do you mean genes encoding transcription factors? In that case answers are not difficult to find. But of course, there is much more to regulation than transcription factors. So the question, which in principle could be answered should be more precise.

The human genome has many different regulatory sequences which are crucial to controlling gene expression. Conservative estimates indicate that these sequences make up 8% of the genome, however extrapolations from the ENCODE project give that 20-40%of the genome is gene regulatory sequence. Some types of non-coding DNA are genetic "switches" that do not encode proteins, but do regulate when and where genes are expressed (called enhancers).

The question was about "regulatory genes". No about "regulatory sequences". I insist, we must be more precise. I consider "genes" sequences that encode a product, be that a protein or RNA. Of course, enhancers, recognition sequences for transcription factors and the like are extremely important but they are not "genes".

I too would like to know the answer, and I think the question is fairly straightforward: a regulatory gene is one that codes for any protein designed to bind to a DNA sequence (promoter, proximal control element, enhancer, silencer, etc.) and thereby regulate the transcription of some other gene. So, what percentage of genes (coding regions of DNA) code for those proteins that are transcription-regulatory in function? When I attempt to search the question on Google, the answers make the same misinterpretation as one of the above, confusing regulatory genes (which code for regulatory proteins) with regulatory sequences (the DNA binding sites of regulatory proteins). Anybody know the answer? What percentage of genes are involved in bureaucracy/administration of other genes' activity?

It is surely known how many DNA binding transcription factorse are encoded in the human genome. I am sure someone must have counted and recorded them. But if you really want to know, take a representative of each type and blast them against the human genome. For example:

Take GATA1

MEFPGLGSLGTSEPLPQFVDPALVSSTPESGVFFPSGPEGLDAAASSTAPSTATAAAAALAYYRDAEAYR HSPVFQVYPLLNCMEGIPGGSPYAGWAYGKTGLYPASTVCPTREDSPPQAVEDLDGKGSTSFLETLKTER LSPDLLTLGPALPSSLPVPNSAYGGPDFSSTFFSPTGSPLNSAAYSSPKLRGTLPLPPCEARECVNCGAT ATPLWRRDRTGHYLCNACGLYHKMNGQNRPLIRPKKRLIVSKRAGTQCTNCQTTTTTLWRRNASGDPVCN ACGLYYKLHQVNRPLTMRKDGIQTRNRKASGKGKKKRGSSLGGTGAAEGPAGGFMVVAGGSGSGNCGEVA SGLTLGPPGTAHLYQGLGPVVLSGPVSHLMPFPGPLLGSPTGSFPTGPMPPTTSTTVVAPLSS

The annswer 6 GATA facrtors, not counting isoforms.

etc