It makes sense to me that DNA is like a config file, while the molecules mentioned in the article are source code, the ‘sensors’ are like a makefile, and the ce

It makes sense to me that DNA is like a config file, while the molecules mentioned in the article are source code, the ‘sensors’ are like a makefile, and the cellular machinery that does the work of reproduction is like the compiler. You need all of the pieces to get a working compiled copy of the program. You can change a lot about the end program by modifying the config file, but you can’t necessarily change the structures themselves. Likewise, you can infer a lot about the end product by examining the config file, but you cannot reproduce it without a compiler, source code, and a makefile. Maybe you can even use a compiler from a distant fork together with source code from another, if enough of the language spec is compatible. Or you could cleverly copy pieces of config files from an unrelated project to create a similar change in your project. Disclaimer: I have no idea what I’m talking about.