3 ms·

I had a thought the other day. It's theoretically possible for a computer virus to lead to a biological virus. Corrupting the data of a computer responsible for

• taylorfinley 7y ago
  Or the converse of this, malicious DNA has been used to exploit sequencing computers. (https://dnasec.cs.washington.edu/ https://dnasec.cs.washington.edu/) I swear I heard Mikki Hypponen talk of a self-replicating version of this that exploited the sequencers to encode the exploit into other sequenced DNA, making this self-replicating, but I can't find a link.
  • mattkrause 7y ago
    The DNA-based attack was a little contrived. They deliberately introduced a vulnerability into a post-processing tool, then generated a DNA sequence that abused it. It’s a creative idea but it would be fairly difficult find and use exploits in the wild. Sequencing has gotten a lot cheaper but it’s still going to cost tens-to-thousands of dollars per attempt, so you can’t do the genomic equivalent of packet stuffing on the cheap. (Nevertheless, it is true that a lot of biology-related code is not great but...)
• bionhoward 7y ago
  Since protein folding is “solved,” you can make pathogen proteins with wildly different DNA to fool algorithms used to identify bioweapons. No bueno because baddies can order nasty germs printed cheap! There are no bug bounties for bio bugs