But the debate should be broadened. Even if we destroy those samples, the war is not over; the smallpox virus has now found a second host. It is not the pig. In fact, it is not even what we think of as a living thing. It is the computer.

This is not some conceptual game. This is real and life-threatening.

If you search online, you can find the sequence for the smallpox genome. It is a word written with the letters A, T, C and G. The word is about 185,000 letters long. It is the word that tells cells to make smallpox viruses. The sequence was stored on a computer in the early 1990s, when a research team led by J. Craig Venter obtained it using a biotechnical process applied to a sample of the virus.

Of course, a word in a computer file cannot kill you. Well, yes and no. In the 1990s, I ran a biotechnology laboratory. In my lab there was a machine much like a soda dispenser, only in this case the reservoirs were filled with chemicals. If I typed in a short word of my choice using the letters A, T, C and G, the machine would squirt one chemical after another into a test tube. When it was done, the test tube would contain trillions of molecules of DNA. Each would look like a necklace, with molecules of adenine, thymine, cytosine and guanine (the building blocks of DNA) strung according to the word I had typed.

At that time, the 10,000-letter sequence of the H.I.V. genome was available online. I contemplated using my machine, together with well-known biotechnical methods, to create, de novo, the H.I.V. genome — an actual molecule identical to that found in H.I.V. viruses living in the wild. I had reason to believe that inserting such a synthetic molecule into a living human cell would cause the cell to manufacture full-blown H.I.V. viruses that could then be transmitted from person to person and cause AIDS.

I decided not to do the experiment, but I began to worry. If I could do it, so could others with high-tech labs.