'Constructing' a living being

Great advances are being made in genetic engineering. How far off are we from constructing a simple living being (say an amoeba) from synthetic, ie non-living components? Is it even theoretically possible?

Genetic engineering is only a tiny part of the problem.

Our current capability is sort of like this. Imagine there is an amazing totally automated factory, that only needs raw metal in ingots, slabs of glass, and bucket-loads of raw plastic pellets to operate. We have the CAD files for a car. If we feed the file into the factory a car pops out. We don’t have any design tools, and we don’t understand a lot of the contents about how the CAD files work, but we have worked out a few useful things. For instance we found the bit of the CAD file that is the driver’s wing mirror. If we remove it from the file, a perfect car, but no wing mirror pops out. Then someone has a go at making a copy of that bit of the file and adding at the end. First time a wreck comes out, with all sorts of mess in the manufacture, half and engine, no tires. But after a few more attempts it is worked out where in the file the copy can go, and a car with two driver’s wing mirrors pops out. Someone else has a CAD file for a different car, and they work out where it’s driver’s wing mirror is in the file, and they try to replace it with the code for the first car’s wing mirror. And eventually they work out how to do it. Eventually someone gets the code for a motorbike, and adds a wing mirror to it. Eventually things get good enough that you can get wing mirrors, number plate holders, hubcaps and side trim on any car you like from any other car. Sometimes this is useful, especially if one car has a really badly designed wing mirror - one that falls off after the first 100 miles.

That is about where we are. We don’t know how to make a CAD design file, only how to mess with existing ones. We don’t really know how most of it works. We don’t really know how to design even a pushbike from scratch. And worst of all, the problem isn’t to make even a push-bike - it is to work out how to make the factory.