Quote:
Originally Posted by ToothSayer
I have no idea what this is about, what the second subject is, or why my name is on this thread. I also don't know more about anything than anyone. I can sometimes reason and evaluate evidence better than others including experts, but that's a really low bar for many fields of expertise.
This one looks like garden variety bio fraud bullshit to me, but again, I know nothing about any of this.
What do you know TS?
DS, I currently work in bioinformatics and have studied ML in grad school. My uninformed and uninterested (no offense) intuition is that given your use of the word “test” and the company name, they are likely over valuing any genetic correlation to severity and underweighting environment variables.
This question, of severity, is actually well suited to ML. The model would use data points like, age weight, smoking status (typical stuff) in addition to a genetic component.
If you want to scalp go for it. But my read is ignorant heard mentality, “dumb” algos, and subpar results if they even manage to go to market.
If you want to invest I’d seek to understand their model and their monetization strategy. I don’t think insurance companies will be chomping at the bit to pay for this one.