Alpha Fold

This is interesting.

How is AI impacting science?

It’s a deep neural network, meaning a hierarchical model, with 93 million parameters learned through training. An amino acid sequence is input, and a 3-dimensional structure is output, together with some error estimates quantifying how confident AlphaFold is in the placement of each amino acid. The basic training data is the protein data bank (PDB), humanity’s repository of the protein structures experimentally determined since the 1970s. At training time, that was 170,000 protein structures (though a small fraction were omitted for technical reasons). The parameters in the AlphaFold network are adjusted using gradient descent to ensure the network outputs the correct structure, given the input.

My worry is that the costs of running this simulation is going to be more than the cost of identification of “protein structures”. This is a backdoor entry into “biological research” where the sweepstakes (given the perverse financial incentives in pharma research) are very high.

I am not a molecular biologist but I am curious about what they think of this tool. It would be an interesting insight into this.