Read large, “scary” story about how the Chinese are so very far ahead of everyone else in the genome sequencing business.
Among the quotes for the fellow running the shop with 157 sequencers:
“if it’s cute, sequence it” and “If it tastes good, sequence it”.
Does this mean that we have enough data to re-create any living object that has been sequenced?
If I had the individual bits of DNA: adenine, guanine, cytosine, and thymine in little stacks, could I (in theory) start stacking them together using the genome sequence data to create the gene of whatever the sequenced specimen was?
In principle, we have nearly all of the information required for any organism with a sequenced genome. Some amount of the information required to get a living cell is not directly encoded in the DNA, rather it is encoded in proteins that bind and package the DNA.
We can also synthesize relatively small DNA molecules through chemical methods. So far the state of the art is around a few thousand bases. In theory you can synthesize an entire genome in small pieces and piece them together with some labor intensive molecular biology techniques.
This has been done exactly once that I am aware of. In 2010,the Venter institude synthesized the genome of a bacteria, inserted it into a related species of bacteria, and ended up with an organism with their entirely synthetic genome. They did this with one of the smallest bacterial genomes, only 1 million base pairs in size, so that the assembly was feasible. (In comparison, the human genome is 3 billion base pairs). And finally, to get this synthetic genome into a working organism, they needed to bootstrap all of the actual working biological processes which require proteins and RNA. This was accomplished by taking a related bacteria, removing its genome, and inserting the synthetic genome.
Essentially what you talk about is possible in principle. But to make one of the simplest possible synthetic organisms, it required a tremendous amount of resources and manpower. I wouldn’t be surprised if the technology was streamlined so that we can make synthetic bacteria with larger genomes, using faster and more reliable methods.
I also expect it will take a decade or three before we can do this with a eukaryotic organism like yeast, let alone anything as complicated as an animal. These have an extra layer of protein-encoded information associated with the DNA genome called the epigenome. We’re gradually “sequencing” the epigenome, but there is no direct and obvious way to synthesize a DNA molecule that is properly packaged up by proteins and RNA with the correct epigenetic information.
Actually, synthesizing yeast from scratch is on ongoing project called Synthetic Yeast 2.0. It is a collaboration between many labs, divided up by chromosomes. I believe their original goal was to do it in five years. I think at least one chromosome is already completed, but I’m not sure. I expect success in well less than a decade. The fun starts after the first success, when they can start making changes to the genome to learn more about biology or to pursue applications.
Incidentally, Craig Venter is now talking about “biological teleportation.” He has a book out called “Life at the Speed of Light” (which I haven’t read). He wants to build automated genome transmitters and receivers (gemome-bots). The transmitter sequences an organism and sends the data to the receiver, which automatically synthesizes the genome. He recently did an experiment with NASA in the Mojave desert to demonstrate some kind of prototype of the “transmitter.”
Nobody will ever live long enough to see the ultimate result of any tinkering. No matter what it is you’re thinking of as “ultimate”, science and technology will reach that point and keep on advancing.
One thing that I think is often overlooked in talk about cloning is that the DNA is only half the story. In a computer analogy, it’s the “software”. The software is useless without the hardware - in this case, a fully functional cell, with the correct complement of proteins and RNAs already loaded and ready to go in the proper proportions, right places, and with all the correct modifications. There are so many ingredients that go into a cell that we’re nowhere near close being able to replicate one.
For living animals, or even very recently extinct animals, this isn’t a big deal, because we can just use cells from living individuals of the same species or a closely related one. But for something that’s long extinct, I suspect we’re going to find this a MUCH bigger obstacle than we think now.
Incidentally, I’d group epigenetics in as a hardware issue rather than software. Epigenetics is all about how and when we read the information, rather than the information itself.
There is also a big jump from sequencing and copying, to fully understanding. Right now we are in the position of having a library full of books in an unknown language. We can copy the words into a new book, we can notice that certain sequences of words seem to go together, and we can even recognize sentences and paragraphs, and may notice that certain words occur in books we recognize as mysteries, and other words occur more commonly in text books. But we can’t actually read the books, much less write our own book in the unknown language.
So while it may be possible in the foreseeable future to recreate a clone of Einstein, there is a big gap between that and creating a woman who looks like Scarlett Johansson but who has Einstein’s intelligence.
I wonder if operations like this were started only to take advantage of patent laws? I know the US Supreme Court had a decision just last year stating that you couldn’t patent a gene itself, but prior to that there was a lot of dispute about whether a genetically engineered organism would belong to the person who engineered it, or to the person who “patented the gene” by being the first one to sequenced it.
If the article you read is more than six months old, or if other countries are still uncertain about biological patents on gene sequences, then this gene sequencing shop could have nothing to do with science and everything to do with profitable legal maneuvers.
Our lack of understanding mostly isn’t at the DNA level, though. Given any sequence of DNA, we can say exactly what protein it’ll produce, and given any protein, we can list all of the possible DNA sequences which will produce that protein. We can even, given enough computer time, reconstruct what shape that protein molecule will be. The tricky part isn’t knowing what proteins are produced; it’s knowing how those many proteins interact to produce complicated traits like intelligence.
In addition, I think one important lesson that is emerging from the last decade or so’s worth of work is that RNA does a hell of a lot more than just convert DNA message into protein sequence (the classic mRNA, rRNA, tRNA pathway). We’re finding entire zoos of different types of small (and large) noncoding RNAs that are doing all sorts of things in the cell. We’re just beginning to scratch the surface of that. There is a longstanding bias toward assuming that RNA is just an intermediary and not important in its own right. I think that’s going to be overturned here in the next decade or two.
Totally agree with both of these. If you take a bacteria and drop in the complete DNA of a guppy, you’re going to get a dead bacterial cell. And vice versa.
Plus, you also need to know the environment that the organism develops in. Not hard to get right for say a common mold spore, but if you drop a fertilized goat egg cell into a chicken egg, or rabbit uterus, you’re at best going to end up with a misshapen, dead, goat embryo (I doubt you’d get to anything you’d call a fetus). For mammals the mother not only provides raw nutrients and shelter, but also various hormones that signal the developing organism.
And even once you’ve got a suitable uterus/egg/wherever for development, you’re also potentially going to need to introduce appropriate commensurate species. Termites would starve to death if they didn’t have the right microbiota in their guts, so there’s no way to build a termite just from the termite’s DNA. Even humans need the right intestinal bacteria to be healthy and digest food effectively.
That’s a weird distinction; I’d call epigenetics the metadata or high-level program flow instructions in the software, rather than the hardware.
The computer analogy breaks down when we start thinking about RNA, of course, which is both hardware, but also self-replicating.
On the other hand, if you clone an extinct animal by finding some modern animal that’s just barely close enough, the offspring you get might be misshapen and otherwise not very well off, but if you then use those first-generation clones for the cell donors and incubators, then the second generation is going to be much closer to the original. Ultimately, all of that complicated cellular machinery and epigenetics is also encoded in the DNA, so if you can get going at all, you’re fine… It’s just that that first step is a doozie.
Not quite all that stuff is really there in the DNA. IIRC, RNA replicates independently from the DNA during cell division, and as smeghead noted above, RNA does lots of stuff beyond helping transcribe DNA into protein. See, for example,
No, that’s not right. RNA is not self-replicating during bacterial or eukaryotic cell division. It is all transcribed from DNA*.
There are some experiments with limited RNA self-replication in hypothetical conditions of abiogenesis. Double-stranded or single-plus-stranded RNA viruses encode a protein that can copy their RNA.
*Except some of the small interfering RNAs, which can be chopped from exogenous double stranded RNA, or be copied from target RNA molecules of other small interfering RNAs. But that’s not directly relevant to cell division. There are always exceptions in biology…
If you were developing the ultimate soldier you wouldn’t start with a biological organism at all. You’d build a robot (or a drone, if you couldn’t figure out how to design an autonomous combat “brain”).