The gene-expression data
The most striking modern GHK claim is that it shifts the activity of thousands of human genes at once, nudging many…
The "resets gene expression" data, read carefully
The most striking modern GHK claim is that it shifts the activity of thousands of human genes at once, nudging many back toward a healthier pattern. The headline numbers are real and come from the Broad Institute Connectivity Map and follow-up analyses.
This unit explains where those numbers come from, what categories of genes move, and the critical caveat that almost all of it was measured in cancer cell lines, not living human tissue. The data is fascinating and genuinely weak as proof of benefit at the same time.
Key terms
The scope of the gene effect
When the review authors ran GHK data from the Connectivity Map through their own analysis, they reported that it changed the expression of roughly a third of the human genes they counted by 50% or more. That is an enormous footprint for one small molecule, and it is the basis of the "resets the genome" headline. Precisely, the figure is 31.2%, against a denominator of 13,424 genes the same paper derives from 22,277 probe sets, and a second review from the same group states the headline as at least 4,000 human genes. The affected genes split roughly 59% pushed up and 41% pushed down. Both figures come from those authors’ own re-analysis at a threshold they chose, not from a primary paper.
AdvancedWhat the "31% of genes" figure does and does not say
The threshold is a magnitude one: a gene counts if its expression moved by 50% or more, regardless of whether that gene matters for health. The authors say plainly that they chose the 50% cutoff, and the gene count rather than probe count comes from a conversion ratio they applied themselves. So the headline measures breadth of activity at a chosen threshold, not benefit. A molecule can be biologically busy across thousands of genes while the net effect on a person is unknown, helpful, or harmful.
A huge gene footprint is striking, not self-explanatory. Changing thousands of genes in a dish does not tell you whether the net effect helps a living person.