5 ms·

I am very inexperienced with biotech. Please correct me if I'm wrong: "synthetic" in this case means the DNA is extracted from one cell, sliced and diced to ge

by umvi 8y ago

I am very inexperienced with biotech. Please correct me if I'm wrong: "synthetic" in this case means the DNA is extracted from one cell, sliced and diced to get a desired genome, and then injected into an existing cell to hijack its offspring to use the new "hacked" DNA. Is that accurate? In my mind "synthetic" means starting from basic chemical compounds and building from the ground up to create a desired molecule. This seems more like if you took a big box of hard drive parts (cannibalized from existing computers), cobbled together a working hard drive, and then stuck it into a computer and then claimed you've "synthesized" a computer.

Pulcinella 8y ago
Correct. Craig Venter’s “innovations” are mostly in marketing and self promotion rather than basic science.
- arrosenberg 8y ago
  Having worked for the man, I can tell you this is a generous description.
  - dnautics 8y ago
    Same (also worked for him). However the internet troll in my heart loves that he basically bilked Wall Street into funding basic science (the human genome project)
TheCraiggers 8y ago
Back when money was tight, I built many a computer from components found in discarded computer cases. Many had lovable quirks, like the one that POST'd in dutch every time and could not be changed. I used to call them Frankenstein computers. While I'd say that would be a far more accurate name in this story, I doubt their marketing department would like that...
tasty_freeze 8y ago
My understanding is that they actually assembled the DNA from small nucleotide sequences, not by taking an actual bacteria and just removing bits. Once they had an assembled chromosome (I believe it was a single ring), the scooped out the dna form a different strain of bacteria, placed their DNA hoop into the empty one, and it reproduced. So, it is still a far way from starting with beakers of the elements, but certainly a bigger accomplishment than slicing some genes out of an existing bacteria.
- entee 8y ago
  This paper appears to be a mix of what you're suggesting, original paper here: http://science.sciencemag.org/content/351/6280/aad6253 http://science.sciencemag.org/content/351/6280/aad6253 They tried paring down a genome to its essential elements and failed. They then combined their designed genome with bits and pieces cut out from the larger, original genome, mixing and matching parts until they found the smallest subset of "parts" needed to run a fully self-replicating cell. The closest CS equivalent is that this was a kind of genetic algorithm with a focus on making a small codebase that minimally accomplishes replication, starting from a set of known-to-work pieces of code. Open questions remaining: - They found 149 genes of unclear function that were required to enable replication, what do those genes do? - If you make a cell this way, how good is it at surviving? In culture, many model systems (yeast, e. coli) outcompete everything in the media by being faster at reproducing themselves (among other strategies). If these cells barely reproduce, other contaminating cells will go faster, making these minimal cells not very good for actual synthetic biology tasks without further modification (maybe they make a toxin while being immune themselves).
  - dnautics 8y ago
    Last I heard many of the GUFs are small integral membrane proteins, so the running theory is that they need to be there (and in quantities) to maintain isotonic homeostasis for the cell. The hard to test but interesting question is whether, though they need to be there, that they are there is more important than what they are.
shawnz 8y ago
The article doesn't really claim that synthetic life was created here. Just that we now have a better understanding of the minimum set of genes necessary for life, which is an important step toward creating synthetic life.
echelon 8y ago
"Synthetic" feels a bit like marketing, but the work is nonetheless cool. Venter's first "synthetic" bacterial cells had their nuclear genomes ejected and then replaced with genomes copied from pre-existing cells. It would be extraordinarily hard to create a cell from scratch without all of the existing biology ("to make an apple pie, first construct the universe".) These genomes contain an abundance of genes, some of which are suspected to be inessential in maintenance of life and proliferation under normal conditions (ie. they signal pathways or encode structural or metabolic components that aren't absolutely critical). Venter's work here is to identify a minimum viable set of genes required to run chosen from an existing, easy to grow bacteria. Such an organism (or set of organisms) would be much easier to study as there is less noise. The biochemistry gets much simpler. To be clear, this isn't the minimum set of genes required to live. It's the minimum set of genes required to run the bacteria it is modeled after. Different experimental parameters may necessitate deriving similar minimal organisms from different species. Or introducing additional genes (eg. resistance genes). Eventually, once these "synthetic" bacteria have been proven, labs will want to order them and conduct their experiments under these much "cleaner" experimental models.
- klyrs 8y ago
  > Such an organism (or set of organisms) would be much easier to study as there is less noise. The biochemistry gets much simpler. An article [1] was posted to HN a while back that gives me pause at that statement. Sometimes, perceived noise is signal and washing it out gets misleading results. Yes I'm sure this will make some experiments easier... but if it's for medical science... I have concerns. [1] https://www.theatlantic.com/science/archive/2019/01/cancer-culture-media-plasmax/579283/ https://www.theatlantic.com/science/archive/2019/01/cancer-c...
chr1 8y ago
There are also people trying to assemble life from simpler molecules that are commonly formed in non-organic processes. This series of presentation by Jack Szostak is rather interesting https://www.youtube.com/watch?v=PqPGOhXoprU&t=1s https://www.youtube.com/watch?v=PqPGOhXoprU&t=1s
antonjs 8y ago
There are different conceptions of what counts as a synthetic cell, but in this work the DNA is indeed synthetic. They designed their desired genome sequence on a computer, and synthesized the DNA by chemically printing it into segments, and then assembling those segments together into a complete genome. There are commercial providers that will synthesize DNA from sequence information [1]. The actual design itself was heavily based on the genome of a natural organism, M. mycoides. They sequenced its genome and built an experimental dataset of which genes were essential and which weren't, then designed and printed a genome using the native sequences of just the essential genes. Work on synthetic cells is now moving in a few directions, including designing the genome from scratch and building cellular or cell-like chassis to 'run' the genome. [2] Even in designing a genome from scratch, however, we would still be using proteins sourced from nature, because our ability to design proteins is still very limited. One way to think of this is that the biological world provides a massive pre-designed parts catalogue, and then we can compose those parts in a synthetic genome. I'm a graduate student working on building a cell [3], happy to answer any other questions you might have. [1] Like Twist, https://twistbioscience.com/ https://twistbioscience.com/ [2] https://www.nature.com/articles/d41586-018-07289-x https://www.nature.com/articles/d41586-018-07289-x [3] http://buildacell.io http://buildacell.io
- dnautics 8y ago
  > They designed their desired genome sequence on a computer, and synthesized the DNA by chemically printing it into segments, and then assembling those segments together into a complete genome. Just to be pedantic, but insightful (hopefully) those 'assembly' stages were in other host organisms (e coli, first, via Gibson assembly, then, yeast using TAR cloning) -- I worked in that lab.
- akvadrako 8y ago
  How long do you think it will be before we have a bio IDE and protein library that's good enough for a programmer to start creating their own living organisms at home?
  - callesgg 8y ago
leggomylibro 8y ago
It might just mean "synthetic" as in "synthetic biology": https://en.wikipedia.org/wiki/Synthetic_biology https://en.wikipedia.org/wiki/Synthetic_biology As a field, it does seem a lot like scrounging through a parts bin for things that fit what you're looking for. But that's fine; we know so little about biological systems, and they are so enormously complex. Plus, the documentation is non-existent and nobody can get in touch with the project's maintainer.

by umvi 8y ago

Pulcinella 8y ago

Correct. Craig Venter’s “innovations” are mostly in marketing and self promotion rather than basic science.

arrosenberg 8y ago
Having worked for the man, I can tell you this is a generous description.
- dnautics 8y ago
  Same (also worked for him). However the internet troll in my heart loves that he basically bilked Wall Street into funding basic science (the human genome project)

TheCraiggers 8y ago

Back when money was tight, I built many a computer from components found in discarded computer cases. Many had lovable quirks, like the one that POST'd in dutch every time and could not be changed. I used to call them Frankenstein computers. While I'd say that would be a far more accurate name in this story, I doubt their marketing department would like that...

tasty_freeze 8y ago

My understanding is that they actually assembled the DNA from small nucleotide sequences, not by taking an actual bacteria and just removing bits. Once they had an assembled chromosome (I believe it was a single ring), the scooped out the dna form a different strain of bacteria, placed their DNA hoop into the empty one, and it reproduced. So, it is still a far way from starting with beakers of the elements, but certainly a bigger accomplishment than slicing some genes out of an existing bacteria.

entee 8y ago
This paper appears to be a mix of what you're suggesting, original paper here: http://science.sciencemag.org/content/351/6280/aad6253 http://science.sciencemag.org/content/351/6280/aad6253 They tried paring down a genome to its essential elements and failed. They then combined their designed genome with bits and pieces cut out from the larger, original genome, mixing and matching parts until they found the smallest subset of "parts" needed to run a fully self-replicating cell. The closest CS equivalent is that this was a kind of genetic algorithm with a focus on making a small codebase that minimally accomplishes replication, starting from a set of known-to-work pieces of code. Open questions remaining: - They found 149 genes of unclear function that were required to enable replication, what do those genes do? - If you make a cell this way, how good is it at surviving? In culture, many model systems (yeast, e. coli) outcompete everything in the media by being faster at reproducing themselves (among other strategies). If these cells barely reproduce, other contaminating cells will go faster, making these minimal cells not very good for actual synthetic biology tasks without further modification (maybe they make a toxin while being immune themselves).
- dnautics 8y ago
  Last I heard many of the GUFs are small integral membrane proteins, so the running theory is that they need to be there (and in quantities) to maintain isotonic homeostasis for the cell. The hard to test but interesting question is whether, though they need to be there, that they are there is more important than what they are.

shawnz 8y ago

The article doesn't really claim that synthetic life was created here. Just that we now have a better understanding of the minimum set of genes necessary for life, which is an important step toward creating synthetic life.

echelon 8y ago

"Synthetic" feels a bit like marketing, but the work is nonetheless cool. Venter's first "synthetic" bacterial cells had their nuclear genomes ejected and then replaced with genomes copied from pre-existing cells. It would be extraordinarily hard to create a cell from scratch without all of the existing biology ("to make an apple pie, first construct the universe".) These genomes contain an abundance of genes, some of which are suspected to be inessential in maintenance of life and proliferation under normal conditions (ie. they signal pathways or encode structural or metabolic components that aren't absolutely critical). Venter's work here is to identify a minimum viable set of genes required to run chosen from an existing, easy to grow bacteria. Such an organism (or set of organisms) would be much easier to study as there is less noise. The biochemistry gets much simpler. To be clear, this isn't the minimum set of genes required to live. It's the minimum set of genes required to run the bacteria it is modeled after. Different experimental parameters may necessitate deriving similar minimal organisms from different species. Or introducing additional genes (eg. resistance genes). Eventually, once these "synthetic" bacteria have been proven, labs will want to order them and conduct their experiments under these much "cleaner" experimental models.

klyrs 8y ago
> Such an organism (or set of organisms) would be much easier to study as there is less noise. The biochemistry gets much simpler. An article [1] was posted to HN a while back that gives me pause at that statement. Sometimes, perceived noise is signal and washing it out gets misleading results. Yes I'm sure this will make some experiments easier... but if it's for medical science... I have concerns. [1] https://www.theatlantic.com/science/archive/2019/01/cancer-culture-media-plasmax/579283/ https://www.theatlantic.com/science/archive/2019/01/cancer-c...

chr1 8y ago

There are also people trying to assemble life from simpler molecules that are commonly formed in non-organic processes. This series of presentation by Jack Szostak is rather interesting https://www.youtube.com/watch?v=PqPGOhXoprU&t=1s https://www.youtube.com/watch?v=PqPGOhXoprU&t=1s

antonjs 8y ago

There are different conceptions of what counts as a synthetic cell, but in this work the DNA is indeed synthetic. They designed their desired genome sequence on a computer, and synthesized the DNA by chemically printing it into segments, and then assembling those segments together into a complete genome. There are commercial providers that will synthesize DNA from sequence information [1]. The actual design itself was heavily based on the genome of a natural organism, M. mycoides. They sequenced its genome and built an experimental dataset of which genes were essential and which weren't, then designed and printed a genome using the native sequences of just the essential genes. Work on synthetic cells is now moving in a few directions, including designing the genome from scratch and building cellular or cell-like chassis to 'run' the genome. [2] Even in designing a genome from scratch, however, we would still be using proteins sourced from nature, because our ability to design proteins is still very limited. One way to think of this is that the biological world provides a massive pre-designed parts catalogue, and then we can compose those parts in a synthetic genome. I'm a graduate student working on building a cell [3], happy to answer any other questions you might have. [1] Like Twist, https://twistbioscience.com/ https://twistbioscience.com/ [2] https://www.nature.com/articles/d41586-018-07289-x https://www.nature.com/articles/d41586-018-07289-x [3] http://buildacell.io http://buildacell.io

dnautics 8y ago
> They designed their desired genome sequence on a computer, and synthesized the DNA by chemically printing it into segments, and then assembling those segments together into a complete genome. Just to be pedantic, but insightful (hopefully) those 'assembly' stages were in other host organisms (e coli, first, via Gibson assembly, then, yeast using TAR cloning) -- I worked in that lab.
akvadrako 8y ago
How long do you think it will be before we have a bio IDE and protein library that's good enough for a programmer to start creating their own living organisms at home?
- callesgg 8y ago

leggomylibro 8y ago

It might just mean "synthetic" as in "synthetic biology": https://en.wikipedia.org/wiki/Synthetic_biology https://en.wikipedia.org/wiki/Synthetic_biology As a field, it does seem a lot like scrounging through a parts bin for things that fit what you're looking for. But that's fine; we know so little about biological systems, and they are so enormously complex. Plus, the documentation is non-existent and nobody can get in touch with the project's maintainer.