Homomorphic encryption, presumably? It's not impossible. But I also think it's overkill. Also, I don't know of good open source software that lets me do the kind of analysis I want even on non-encrypted data.
That would not be very comforting, this would mean that even encrypted, we can know stuff from the encrypted form, which kinds of defeat the purpose unless I'm missing something.
That being said, scientists can implement their own protocols, and use whatever technique they want. For example: https://github.com/securegenomics/protocol-alzheimers-sensit....
It's that our platform makes federated computing + homomorphic encryption analysis easy, but protocols are customizable.