This is a test version of Biostars. For the public version, visit https://www.biostars.org.
Average Number Of Micro Target Sites For Human Genes

Dear community,

Since some of you are very experienced scientist, I would like to ask for some expertise: Can you tell me how many miR binding sites genes have on average? My research totally depends on number of miR binding sites, and I took targetScan's estimates, which gives me around 20~30 on average. But my boss thinks that number is too many for genes, and wants me to prove the number using literature.
And he wants a direct evidence, like a distribution figure of the miR target site number, for example.

But I can't find one, after trying hard for a couple of weeks. If you have ever seen this kind of literature, can you please let me know of them and save me? I know it can read as a personal request, but please understand my situation. 

Help will be greatly appreciated.

binding
• 3,274 views • link
updated by Michael Dondrup Michael Dondrup • written by thejustpark thejustpark

Btw.: did you google? "validated miRNA target sites", but this search shouldn't take weeks ;)

• 0 views • link
written by Michael Dondrup Michael Dondrup

I know. But I have to find direct evidence like a distribution figure as I mentioned. I may have to generate this myself according to literature.

But thanks for the your reply.

• 0 views • link
written by thejustpark thejustpark

1 answer

The main line of controversy with your boss is possibly between "computationally predicted" and "experimentally validated". You gave the predicted range as 20-30, and your boss argued with the validated figures. For example the statistics page of mirTARBase yields:

Species    No. of miRNA-target interactions (MTIs)    No. of miRNAs    No. of target genes ....    
Human    38113    587    12194     ...     |  average  3.125554 per >targeted< gene
Total    51,460    1,232    17,520    ...   |  average 2.937215 per >targeted< gene

This is the lower bound, so the real number might be somewhere between 3 to 30 (or above) on average. Unless there is or you make also a true negative list of falsified predictions, it is hard to estimate the real figures.

• 237 views • link
written by Michael Dondrup Michael Dondrup

Hi thejustpark , Both the numbers, 20-30 or 200-300 is what has been reported in literature. If you mostly focus on stringent (8mer/7mer), 20-30 is about the right number. Whenever 6mers are included, the number rises to about 200-300 genes. High-throughput sequencing of dicer mutant rna show that it is possible for one mirna to target 200-300genes, but the mrna fold-change is often around 1.5fold-change. Please follow David Bartel's paper over the last few years. You can also download targets predicted using high-throughout CliP-seq.

• 0 views • link
written by Chirag Nepal Chirag Nepal

Dear Chirag,

Thanks for this comment. Can you please let me know which literature reports 200~300? I found some papers with the number, but they were not direct enough.

• 0 views • link
written by thejustpark thejustpark

Mammalian microRNAs predominantly act to decrease target mRNA levels

• 0 views • link
written by Chirag Nepal Chirag Nepal

It seems this paper talks about the number of genes one miR (miR-155) targets, while my question was about the number of miR target sites per gene. But thank you for letting me know.

• 0 views • link
written by thejustpark thejustpark

Log in to answer this question.

More posts like this