Sequencing Genomes

It's not a podcast, but I thought people might be interested in learning how bacterial genomes are sequenced. I am participating in the sequencing of two of my strains, at the moment.

What this entails, apparently, is driving to a nearby city, to the university there, which has a 454 pyrosequencing facility.

Pyrosequencing is a rockin' cool concept. It sure takes a while to set up though (about two weeks, it seems). It goes like this:

1. Isolate DNA from the bacteria.

2. Cut this DNA up into little pieces, and attach specific sequences to each end.

These two parts took all day Monday. Then we gave the results to another facility to make sure the pieces were the right size. That should take the rest of the week, because they're busy with other stuff too, I guess.

Next week:

3. Emulsion PCR: This means that the little pieces of DNA are put in an emulsion of oil and water, such that each piece is in a little water droplet with a tiny bead and all the chemicals and enzymes needed to copy the piece of DNA many times. Each copy is attached to the same bead, millions of copies on the same bead. This is apparently a boring step that also takes all day.

4. The beads are taken out of the emulsion and everything but the beads is discarded. This will be next Wednesday, I think, and I get to help.

5. Not sure what happens Thursday, but on Friday I get to help set up the plate, which looks something like this:


This is a (very expensive, about $1000) plate that contains many many tiny wells in the yellow areas, each well only big enough for one of the tiny beads to fit. So one bead goes in each well, and the plate goes into the sequencing machine:


Which is shiny and probably very expensive also, tens or even hundreds or thousands of thousands of dollars. It has this sticker on the back:

Which means, I guess, don't throw it in the trash. Good advice. I found that rather hilarious.

So what this machine does, is it washes the plate with one kind of base at a time.
Note: DNA is made of four bases, mostly: adenine (A), cytosine (C), guanine (G), and thymine (T). These make up the unique sequence of each DNA molecule.
So the machine washes the plate with one base, and if that base is what comes next in the sequence of the piece of DNA in a given well, that base will be incorporated into a copy of the sequence, and the bead will give off a flash of light, which the machine will detect. Then the plate is washed with another kind of base, and different wells flash. The machine records the flashes, and this is how the sequence of the DNA on each bead is determined.

So that's how a genome is sequenced. Actually, one plate can hold enough little pieces to make up more than one bacterial genome, which is good because you have to use the whole plate at once, and it's expensive ($5000-6000 I think) so you don't want to have to do more than one if you can avoid it.

After I get the sequences of all the little pieces of DNA, I still have a lot of work to do. I have to figure out how they fit together, for one thing... thousands of overlapping pieces, each of which make up about 0.014% of the whole genome. Fortunately there is software to help with that, and I have a related genome to compare it to. And then there is other software to figure out which parts of the genome are most likely to be genes, and then other programs to figure out what the products and functions of those genes might be. I am supposed to be finished with all of that by the end of the summer. It should be fun, I guess.

So that's how it's done. Genomes are sequenced this way, as are single genes from many different organisms in an environment, to see what's present. It's a good way to sequence a whole lot of DNA all at once. Hope you found it interesting.