A brief excerpt from a YouTube video (created and uploaded by Joy Kong, MD) where two speakers discuss how biological noise may affect age-prediction results and question how other companies calculate biological age. The exchange raises concerns about the impact of noise on gene-expression–based aging metrics and the limitations of simple linear regression approaches.
So you're saying these biological noises, you think that's affecting.
The results.
That these other age companies are testing, how they're calculating?
So it's not the noise that they're using, it's the model that they're using. It's having a lot of fluctuation when you're using a linear model, even if you're having a slight adjustment, that's having a lot of huge changes on the output of the prediction. So what we decided to do is we're not going to be looking at the linear regression model. What we are doing is we see what is the expression noise level. So what we discovered is that aging is very similar to entropy, which means in a younger state, you have a very tightly controlled gene expression, which is the noise is very low. But as we are aging, the fluctuation of the noise, which is the randomness of the expression, so starts to be expressing even more. So it's starting to be having bugs in your gene expression. So that's what we are referring to as the biological noise. We test what is the noise level in your body.
