The biological world is very different from the world of man-made devices and you cannot expect to easily engineer, and even more so, to "program" things using DNA the way you build something from NAND gates or computer instructions - in engineering we strive very hard to make parts independent and abstract away low-level details, in biology everything is interacting all the time (even physically the molecules clash hundreds of times per second) and every low-level physical/chemical effect might be used to serve some purpose. Evolutionary "design" does not have any limits to the intellectual complexity of its results like our limited human intelligence forces us to have in our "purposeful" design process.
Even predicting the outcome of a single gene being expressed is very complicated, protein folding is now a separate field putting our knowledge of physics, chemistry, biology and computer science to a test, and what to say about predicting the interactions of those proteins inside the human body. I wish people would take a good university-level biology 101 course before making nonsensical simplifications and spreading those as "science news".
By the way, I do not doubt this is an interesting initiative and that some useful things will come out of it, I just strongly oppose the notion that we can all now "play with nature’s design".
I totally agree with this! I wish we could "play with nature's design" like we can program computers. But alas, our technology is woefully inadequate for the task, and our understanding of the natural world is still sadly primitive. What is needed is a generation of people willing to commit themselves to basic science research.
Working in basic science is perhaps not as glamorous as working as an entrepreneur, a programmer, or whatever else. But over the course of a lifetime, the work is infinitely more satisfying.
Curious what field and capacity you're working in basic science. I'm considering pivoting out of industry. Studied physics undergrad but quite frankly couldn't cut it to do so post-grad. However I do wonder if another field would be doable for me... but without prereqs I'm not sure how to get onto that path now (I'm not a coder, getting really into bio/chem/neuro/genetics). Been out of uni 6 yrs
You can go into biology with no background. I went from physics and math as an undergrad to a PhD in biology. The first six months were rough, but if you really cut yourself off from your previous field and make yourself understand how the thought process in biology works, you can pick it up fairly quickly. Actually, I suspect that an undergraduate degree in biology might be a liability, since you will have memorized so many things that were simplifications or just outright false.
On the other hand, know that the field is incredibly political and that the old joke about the average IQ of both physics and biology going up when Delbruck switched from the former to the latter is true.
Well great to hear you've made a similar transition, and very cool to think that some of the bias coming from traditional lower-level bio training could be looked at as a liability.
I wonder if I can somehow find a way to pivot into bio without going through a university. Would love to somehow get there through industry, perhaps through some of the genome sequencing outfits which have some crossover with the semiconductor industry.
Hadn't heard the Delbruck joke but hah good to hear, if I understand you correctly, that in a sense it'd be easier to swim in bio if I was sinking in physics.
"our understanding of the natural world is still sadly primitive" You are talking for yourself ? For humanity ? For scientists you never met ? I honestly don't know who you are and how well suited are you (you're a specialist in bioengineering ?) to tell us how much "we" know or we don't. I'm just a software engineer with some basic understanding of genetic engineering, but my wife is a genetic technician in a lab, and it's amazing what operations they performed on a daily basis as part of their university classes and recently in research projects in real labs (and this didn't start yesterday or only with my wife's generation). So I'm very tempted to believe that you just woke up speaking about some limitations you don't personally know and it's not even in your area of expertise. This does not mean that today "everything" is possible or that we achieved the grand mastery of DNA. It's just that for the "right people" the understanding of the natural world is more advanced than you have ever dreamed (unless you are one of the "right people").
Computation chemistry has been going on for decades. Attempts at modeling how molecules behave is still quite primitive. We're talking about modelling the behavior of a object that is comprised of a few dozen atoms. That's it, pretty simple right? We'll the models aren't that good at predicting molecular behavior.
Let's move up a step now. Computation chemistry is used heavily by the drug industry. Get an x-ray structure of a protein (maybe a few hundred to a few thousand atoms) and see if it binds to a drug. Wow, now it's getting complicated. How successful is it? Not very. I can remember a computational chemist saying "oh hey, the model say if you replace X with Y, you'll increase binding by 10x". So we try and guess what? The binding was worse.
Now we move up to a biological system. Now we have hundreds (if not thousands) of proteins floating in a matrix of water and ions. We have a DNA strands of millions of base pairs, of which maybe 10% we actually know what they do. We also have small signalling molecules that do something we understand, but probably also do 10 other things we have no idea about.
It is very impressive how far biological "design" (genomics) has come so far, but right now the tools are incredibly blunt and the analysis is incredibly crude. I have no doubt our understanding will improve immensely over the coming decades, but I would guess we understand less than 1% of what's going on inside of complex living organisms.
You see, my feeling is that trying to guess the progress of genetics by extrapolating the complexity and insecurity of the computation chemistry is wrong. As far as I understood, genetics today is a lot about (but not only) identifying which genes (portions of DNA) are responsible of which phenotype (en.wikipedia.org/wiki/Phenotype).Therefore a lot of resources are and were allocated to create a dictionary with genes as keys and phenotype as values. This dictionary is being populated at a quite fast pace and this combined with the possibility to take genes from some organisms and implant them in the DNA of a cell of other organisms and see the resulting phenotype is already a great achievement (imagine undergrads cutting and pasting DNA daily). They are not inventing new proteins and worry that they will not "bind" enough. They just take the DNA known to produce proteins in some organisms and place it in other organisms and suddenly proteins which have a known effect in different organisms, appear in a new organism. Yes, there is a long way from here to engineering genes that will produce and deliver a medicine inside an organism but I wouldn't call this primitive at all. Sorry for the simplification and possible errors.
OK, I see your perspective now. I agree that our understanding of how genes encode for proteins is well developed, as are our techniques for "transplanting" a gene from one organism to another.
What we have very little handle on is gene regulation. All those "non-coding" genes that scientists used to think were junk? They are actually used to control gene transcription.
Controlling this is infinitely easier in a simple organism like a hookworm, but the complexities of in human borders on obscene.
My prediction: We can find a subset of proteins such that they do not interfere with each other, and still large enough that they can perform useful functions.
It is possible to write threaded software in a way that everything interacts with everything and it is almost impossible to make out how anything works. That's why we don't. The halting problem never stopped us from writing software.
Proteins are very "sticky". Even proteins with weakly predicted interactions exhibit significant effects upon one another when all-atom simulations are performed - this is actually a major stumbling block in getting cellular simulation right.
The most complicated piece of multithreaded software yet devised by humans does not compare in complexity to the transcription, translation, and interaction events occurring in a typical human cell. The "DNA as software" metaphor is just that, a metaphor - it is an exceedingly poor model. Cellular systems are so quantitatively enormous and convoluted (yet not chaotic!) that we have to compare them to the most complicated designs our species has recently engineered just to begin to get our heads around the problem.
I haven't kept up with the state of the art in the past two years, but I remember an experiment running a ~10k node cluster for several weeks being able to successfully simulate only the cytoplasm of a cell - and 1/1000th of its overall volume at that. And the proteins were all modeled as spheres. I'm sure the art has advanced, but that is orders of magnitude away.
My prediction: We can find a subset of proteins such that they do not interfere with each other, and still large enough that they can perform useful functions.
The tiny, tiny amount of bioinformatics knowledge I have makes me think the probability of your prediction is ~ 0.00000000000001%. For non-trivial values of "useful". :)
Can you tell me what experience or knowledge lead you to make that prediction?
Biological systems are highly coupled but also very "fault" tolerant if manipulated at the right pivot points. That's why you can move a fly leg from the torso to the head by just manipulating a few transcription factors.
Now I feel bad, maybe I should have voiced it in a way that sounded less sure. I base it on two things. 1. Similar genes do similar things in different species. 2. There are an incredible amount of possible proteins.
While I might agree with your general point that biology is not software, I think you might be being a little too pessimistic for the field as a whole. You don't need to predict the "outcome" as you define it. You just have to get the function you want. Plenty of people, even high school students and undergraduates, are "playing with nature's design."
I am not necessarily pessimistic, I do believe smart people will be able to work those things out and some maybe even will put them to good use, I am just an opponent of oversimplifying problems and sceptical of metaphors. You say: You just have to get the function you want, I would put it more as You have to get just the function you want and I think this isn't as simple, especially when those artificial organisms start interacting with environments more complex than a test tube e.g. the human body. In this case, outside of the laboratory, I think it would be valuable to know what we are doing and not "play".
This probably depends a lot on the context you would like to use those things in, but in general there are lots of warnings about potential unexpected interactions in descriptions of some of the parts and the practical projects seem to have lots of safety precautions too.
Don't deceive yourself into thinking that you can program your way into biological research. Biological research has very little to do with apps/programming/web-development.
Synthetic biology requires an understanding of chemistry, molecular biology, and genetic structure.
DNA manipulation would be orders of magnitude more difficult and time consuming with out computers as would anything involving chemistry, molecular biology and understanding genetic structure.
Apps and programming/automation have everything to do with the future of the field, its not just test tubes.
True, but from what I've seen biologist for the most part can get along just fine with only a basic knowledge of REGEX and spreadsheet software.
The biologist I've worked with tend to be a resourceful bunch, capable of solving difficult problems with simple tools.
EDIT: Granted, there is a growing need for data-scientist who can help sort through mountains of data for relevant information. But I don't think there is much need for pre-built software solutions. Each lab faces unique problems, and their software development needs, if any, are unique.
What if I have that knowledge (in addition to software engineering) and would like to contribute in some sort of open-source capacity? Know of any other projects I could look into? :)
How do they prevent the E. Chromi bacteria from eventually being displaced by other non color coding poo bacteria? It seems to me that even if the E. Chromi did work as advertised, that the bacteria would quickly (within days) disappear from your intestinal track.
If anything, I would suspect that the E. Chromi would be at an evolutionary disadvantage in the gut. This is because the E. Chromi would be expending effort trying to color code poo, while their neighboring bacteria would just be focused on digesting and reproduction. Based on the principles of evolution, I would expect the E. Chromi bacteria to disappear entirely from the gut with a few thousand generations of bacteria.
Very important point. E. coli describes a range of organisms that share about 14% of their genome. For comparison, all of primates share over 99%. The strains everyone uses for genetic engineering are derived from strains isolated many decades ago and are incapable of colonizing hosts. They've also been selected over the years to be hysterical (they over respond to any stimulus, since when you're picking colonies to study response, you select the bright, clear one...and if you do it again and again over the years, you select for hysteria).
Poking around, I couldn't find the strain they used. They might have engineered a gut isolate. They might not. Doesn't appear to say anywhere.
Past participant of iGEM (International Genetically Engineered Machine, the competition mentioned in the article) here. Infuriatingly, this writeup fails to mention that the E. chromi (and all the other projects) was carried out by undergraduate students (this is the team: http://2009.igem.org/Team:Cambridge/Team), perhaps because it makes it sound less trustworthy and sellable. According to the same website, the "designer Daisy Ginsberg" was involved peripherally, at best.
Hi, My name is Omri and I'm the founder of Genome Compiler (genomecompiler.com) - we built a free app for designing and ordering of synthetic DNA - check it out!
Very cool idea! I think you should think carefully about the use cases.
Is it for someone who wants to take a gene from one organism, move it to another and then order that organism?
Is it for people wanting to manually tweak genes to improve efficiency, slightly change function? What kinds of manual tweaks?
Is it for someone who wants to understand how an organism works?
When /I/ look at it, I think it could benefit from being more abstract, but I am not your target audience, so take this with a grain of salt. I don't think looking at individual base pairs is useful. Amino acids might be, but are still too emphasized in the interface. The most abstract view, showing genes in order is not abstract enough. I want to see genes grouped by function (e.g. reproductory system, energy production, acidity regulation etc). For each gene I want to see:
a) High level description of function (already there, but in a single line text field. Give more screen estate).
b) What activates the gene? i) Directly (e.g high concentration of Na+) ii) Indirectly (e.g. genes a, b, c) iii) Very Indirectly (environmental stress)
c) As exact specification as possible of what happens when it is activated: i) What does it activate in turn? ii) What does it catalyze?
Others have already said this, but let's get it out in the open once and for all:
Biochemistry has no relation to computation.
Neuroscience has no relation to computation.
Anatomy has no relation to computation.
Ecology has no relation to computation.
Someone using an computational image in any of these fields is trying to impart a sense of familiarity for his audience. If you actually want to learn any of these fields, you need to build your thought processes from scratch. No part of you works like a computer.
This may be nit-picking but in their diagram the color chart, portrayed visually as a venn-diagram, is rather misleading. It seems to imply that if you have colitis and worms, you have colorectal cancer. And if you have everything, you're A-OK. Really, a color wheel would have been a far better choice of diagram.
http://partsregistry.org/Catalog
The biological world is very different from the world of man-made devices and you cannot expect to easily engineer, and even more so, to "program" things using DNA the way you build something from NAND gates or computer instructions - in engineering we strive very hard to make parts independent and abstract away low-level details, in biology everything is interacting all the time (even physically the molecules clash hundreds of times per second) and every low-level physical/chemical effect might be used to serve some purpose. Evolutionary "design" does not have any limits to the intellectual complexity of its results like our limited human intelligence forces us to have in our "purposeful" design process.
Even predicting the outcome of a single gene being expressed is very complicated, protein folding is now a separate field putting our knowledge of physics, chemistry, biology and computer science to a test, and what to say about predicting the interactions of those proteins inside the human body. I wish people would take a good university-level biology 101 course before making nonsensical simplifications and spreading those as "science news".
By the way, I do not doubt this is an interesting initiative and that some useful things will come out of it, I just strongly oppose the notion that we can all now "play with nature’s design".