>What you describe involves having libraries of DNA from another genome, characterized, and the tools to cut it, no? I assume an organism wouldn't have this available Yes and no. In a laboratory, you harvest a DNA plasmid, use a [purified restriction enzyme](https://en.wikipedia.org/wiki/Restriction_enzyme) to create the ends of the DNA you seek to hybridize, then use polymerase to insert your gene, then insert that plasmid into the host cell. But those enzymes exist in nature... but yes some wild cell out in nature isn't going to just encounter, idk, a plastic straw and think, "Oh, I need a protein that can cut PVC, I'll just build a DNA sequence that codes for a protein that can do that!" But certainly genes are accidentally spliced in nature. >Or do you actually mean just composing new sequences from scratch to do some desired thing? Usually the sequence to code the protein is actually built from scratch. DNA/RNA synthesizers can be had for ~$10k or so, and I think there are synthesizers out there than can produce entire genomes (been out of that game for over a decade, but I would suspect they exist). In my student research days ~20 years ago it was just: translate the AA sequence to an RNA sequence, translate that to DNA (avoiding thymine dimers and other rules), then drop that off at the "synthesis shop" and come back in a day or however long it took them to get to you (sometimes same day). Like dropping off fabrication specs at a machine shop... kinda nuts looking back on it.