> New CRISPR-based map(ping process) ties every human gene to its function
Maybe it's because I'm not in that industry, but I was looking for a graphical gizmo that I could click on a gene and see functions but they mean "map" in its functional programming term
Although clicking through the first link does say "Interactive Website under construction..." so maybe this was just submitted too early or something
To understand its function, it has its own wikipedia page(!) https://en.wikipedia.org/wiki/C7orf26
It's one of the remaining proteins whose "function" (to the extent that proteins can be said to have a "function") has not been reliably determined.
as you can see, they don't really know what it does:
"Probable component of the Integrator (INT) complex, a complex involved in the small nuclear RNAs (snRNA) U1 and U2 transcription and in their 3'-box-dependent processing.".
In general, data presentation in biology is a pretty mixed bag. The field never attracted the level of UX investment that you see at ad-driven companies.
That’s true, software for biology tends to be terribly, almost comically bad, with one-off file formats, brittle data interchange, and impossible-to-maintain code being the norm. With user interface and ergonomics being the most neglected aspect. Why do you think that is?
Surely there is plenty of money in biology these days to hire a good designers to design good user experiences. Surely better user experience for biology software would lead to better understanding of biological systems and better outcomes in bioengineering.
Where are the polished, powerful design tools for biology like those that exist for other fields like online advertising that routinely process and distill huge amounts of lightly-structured data?
User interfaces for biology have drastically improved over the last 10 years.
Domain-specific tools like genome browsers, protein viewers, or phylogenetic explorers [1-3] almost all look and feel a lot better than they did in 2012.
The biggest exception here is UCSC Genome Browser, which has an old-school design and web technology stack. That said, it's steadily added features over the years, has substantially sleekened UX in its periphery, and remains widely used.
There are also bespoke visual design resources for biology applications that are good and getting better, like BioRender and PhyloPic [4-5]. There are multi-tiered packages like Dash Bio that wrap biology components together [6]. There's a Blender biology community, too!
Software quality improves once you move to older well-established problems, where developers who are not actively involved in research can contribute.
If you work with cutting-edge problems, the job of a programmer is turning vaguely expressed biological questions into software. The first attempts will inevitably fail, as they try to solve the wrong problem in the wrong way. Technical debt will accumulate at an incredible pace, as your understanding of the problem improves and the software becomes something very different from what you originally imagined.
I've been trying to break into biotech for years...short answer is they want biologists, not programmers, and don't appreciate skillsets that aren't already secondary to PHD industry-relevant experience.
The reason why is it is easier to train a biologist to become a programmer than it is to train a programmer to become a biologist. Sure the quality of biologist-turned-programmer code will often not be great, but it will usually answer the question asked.
Not in my experience. The reason is because of funding. NIH grants are focused on biological discoveries and not so much around infrastructure. So you can get funding to build a tool, but not so much to maintain it. The downstream effect is that tools and websites are stuck in the era in which they were created, databases are not updated, and tools are broken because the grad student left and now there’s no one to respond to issues.
The argument I've heard for such domain specific software having poor UIs is refined UIs only matter for a the first third "10,000 hours". After becoming an expert you simply become used to the software. Anybody professionally using the software would only see a _loss_ in productivity - even if the UI changes for the better.
it seems like that is a whole other skillset vs. biology graduate students...I know the Allen Institute employs data vis people to write visualisation tools like this in d3/react or whatever the web api du jour is, but they are separate from the science folks
Wait do we even know that every human gene has a 1:1 mapping to a function in the body? I think the code analogy to this would be trying to map "C++ if statements" to a particular feature in your product. I'd expect something as complicated as the genome would have a much richer and complicated interaction with biological feature expression.
I'm probably simply misunderstanding; maybe Weissman's data is a 1:N mapping?
> Wait do we even know that every human gene has a 1:1 mapping to a function in the body?
When it comes to biology, the answer is typically: It's complicated. I'll give two examples, melanin and the gene for vitamin-C.
The process to make melanin is a bit complex, but suffice to say the chemical is produced. From there, it is used in at least two way you are probably familiar with: sleep and tanning. Our bodies uses melanin production to help regulate our sleep cycle. Also, through the quirks of evolution, we use it to help stop UV radiation and keep our skin safe. At the end of the day, a complex process that includes a few genes maps to multiple functions.
The gene for vitamin-C production is another example. In humans, some primates, and guinea pigs, this one gene is mutated. As such, we can't produce vitamin-C internally and we have to get it from our diet. This single gene had a function in our ancestors, but no longer does. I think you can argue that it's lack of function is a psuedo-function for our ancestors to eat more fruits, but I don't really buy that. Evolutionary, we would be better off with the non-mutated gene than with what we got now (ask any scurvy survivor or Inuit).
Generally, I'd say that most of the genes in our code are more like the mutated vitamin-C gene than the melanin ones. They had specific functions sometime in our history, but no longer do. Kinda like a vestigial organ or the eyes of a cave fish.
This kind of shallow presentation is common for genetics; compare reporting genetic traits by SNPs and implying the SNPs "cause" them, which is like diffing two versions of a program and saying X new feature happens because there's a letter 'n' in one program and not the other.
There’s not any type of mapping and this is what makes biology so difficult to study and understand. One gene can affect anywhere from zero other genes to every single gene. We also need to account for effects at every level of the central dogma of biology. Knockout of a gene can affect genetics, RNA, protein, and much more we don’t even know about yet. The other underestimated complicating factor is that biology is very nonlinear. A 10% increase in gene expression could lead to no resulting effect, whereas an 11% increase in expression could upregulate downstream genes by 1000x.
I remember going to a really interesting talk in which the proposal was to represent the transcriptome as a probabilistic generative model. That felt like getting something close to an appreciation of the complexity.
I think mpeg encoding would be a better demonstrative than code...
Where data is lost, details are lost, but the end result still typically renders into something recognizable. If you lose an iframe, you end up with a more serious deformity. Whereas code either does exactly what it says or does nothing at all, and knocking out a single statement is almost certain to break everything.
There's an interview with Ken Thompson from about 2008 where he says that most of the work in CS has already been done, and he's advising his son to go into biology.
I feel like CRISPR is the transistor of the 21st century.
CRISPR is more like a machine to make large-node-size integrated circuits.
Somebody else has to design the circuit, make it manufacturable, and integrate the circuit with a whole bunch of other hardware.
I work in biotech at a company that is one of the few golden geese that lays 2-3 successful drugs with no competitors every few years. I have 30+ years of experience (deep experience) in machine learning, biology, and computer science.
We are so far behind where we could be, in terms of turning biology into technology, that's almost shameful. Every day I see another system that says it can generate 10 times the data of the previous machine, but the actual amount of knowledge we are extracting for all that data collection is growing logarithmically. This is because for a long time biology has greatly underfunded computing and data.
The one great shining light is AlphaFold. AF2 finally demonstrated to a wide range of scientists across many domains that a really great team using techniques that are barely known outside of FAAMG can work with some long-term experts to move a metric (quality of predicted protein structures compared to golden data) substantially further and faster than even the most wildly optimistic predicted. Not only that, some of the techniques they used didn't even exist several years ago (transformers, jax, various graph learning systems), and the work was replicated externally once the leading academic team had a hint of the direction to go in.
To me, nothing about what I said is surprising to me; I predicted these outcomes a long time ago. Most of the reasons that it comes slower than it could are combinations of culture, incentive, morals/ethics, politics, innovator's dilemmas and a hundred different bottlenecks. Recently, the challenge has been that most of the really smart computational biologists disappear into FAAMG and don't contribute back the things they learn there to research.
We're all waiting for that next moment when the cross product of Genentech and Isomorphic Labs announces that they have a computational model that can do end to end prediction of drug, from initial disease target to FDA approval post-phase III trial. That's been the dream for some time but we're nowhere near it still, and it remains to be seen whether some group can conjure all the necessary bits to solve the remaining underlying problems associated with that "far beyond NP-hard problem"
I studied CS and now work in software systems for biomedical research. It's difficult to overstate how different the fields are, so I don't entirely agree with this statement. But I do agree there are going to be lots and lots of huge discoveries in biology in the 21st century.
The main difference is that CS attempts to generate and study complex systems built from well-understood components, whereas biology attempts to understand and manipulate systems that evolved naturally over eons.
Imagine dropping a fully functional internet-connected Google Home Hub into 1960-era humanity and asking them to figure out how it works so they modify it to sound like Walter Cronkite. There are thousands of problems on this order of complexity in biology. It's wild.
Same here. The more I've progressed in CS, the more dissatisfied I am. Outside of creating algorithms that vie for constant user attention (the basic business model of FAANG), I don't see any fruitful application of my skills. I'd much rather move towards the domains where my knowledge of data, systems, and algorithms could be better utilized (medicine, Genomics, structural engineering, governance etc).
I've done the physics->neuro leap, so I may be of some use here.
The path is pretty clear, but takes time. Essentially, you need to go back to school and learn biology.
Fortunately, many grad programs in the US are desperate for people that want to be trained as biologists but have relevant skills in other areas like CS. So skip going back to undergrad and just apply to grad programs.
Unfortunately, that means you have to join the Ivory Tower's horrible system for a while. A 'good' tactic is to get into a PhD program where you'll be paid, learn everything, get your MS, and then quit the program after ~3 years with a free MS. Fair warning, the learning will be absolutely horrible and you'll be on the bubble of being kicked out; it really is that much info you're trying to digest in such a short time period. But if you're not worried about scholarships and grades, then that's fine. Your PI will hate you, but then again they hate everyone, so it's a wash.
One thing to be clear about though, jobs in biotech are much less well paid than in CS. You're looking at a 1/3rd to 1/4th salary decrease for pure bio jobs as compared to programmer jobs. Even leveraging your coding skills for biotech companies is going to be tough; you'll be pigeon holed into either a lab role or a coder role. The true blended roles are very rare. So much so as that you'll essentially have to start your own company, or be the heart of any company your join. So, good money there, but huge pressures.
Having done the physics -> neuro leap, it's pretty tough.
You have to learn a whole new set of fields and new ways of thinking. That takes time. To be 'good' at genomics, you kinda need to know how the genes are implemented in the various model organisms. Which means you need to know the relevant biology, biochemistry, chemistry, and physics of the situations. That's, essentially, an entire undergrad education. Then, you get to do the actual work, which takes about 1.5 years of study, so most of a masters degree. Then you can start really doing the work.
For me, the first big realization coming from physics was that these little yeast cells and zebrafish aren't just little machines of quantum chemistry. They really are alive, even down to the cellular level, and they are studying you too. There were hundreds of such insights.
3.7 billion years of refactoring has kept it pretty clean and functional. We'll need to do a shit ton of unit and integration testing before we commit changes.
There is a lot of pruning that occurs, evolutionarily speaking,
and a lot of what was thought to be "useless" genome has been discovered to be conserved over generations, and that there is use for that part of the genome.
The problem with this is that biology will likely end up dominated by China due to a willingness to conduct experiments that are otherwise non-viable in most countries.
Yeah, it feels like most fields are stagnated except for biology and neuroscience. I am a postdoc right now but have considered seriously switching fields to work on something exciting.
I'm still holding out for photonics and other optical-analog computers (where are my instantaneous trig co-processors?) but that does sound like good advice
About 18-24months ago I went through a phase of listening to geneticists talks/conferences/podcasts for an hour or two a day during long runs, it’s so far outside my wheelhouse I’m probably mixing things up, but I thought I recalled cutting edge experiments using synthetic cells to create artificial life (a worm perhaps with a relatively simple DNA, maybe even a modified DNA further simplifying the genome to the furthest extent possible still resulting in life) with one of the goals of understanding the exact functions of all the genes in this “simple” DNA. Again I’m probably mixing up multiple discussions and studies into one, but I would have been very surprised if the function of every single gene in the human genome was known and understood, as suggested by the title.
I worked with mycoplasma genitalium which is a "minimal" organism- an extremely small number of genes, nearly all of which appear to be absolutely required for viability. It's sort of a unit test for model biology, except it grows so slowly it's more like an integration test in terms of performance.
This expert from the first link is very likely what I was poorly trying to regurgitate:
> Mgen still has the smallest genome of any known (naturally occurring) self-replicating organism and thus is often the organism of choice in minimal genome research. The synthetic genome of Mgen named Mycoplasma genitalium JCVI-1.0 (after the research centre, J. Craig Venter Institute, where it was synthesised) was produced in 2008, becoming the first organism with a synthetic genome.
> The work in this area is quite extraordinary, but typically gets much less attention than anything that works with human genomes.
In fairness laymen like me would just get us all mixed up with a worm genome anyway ;). In my defense I’m just a lawyer that likes to listen to foreign topics I find interesting while I run, but it is nice to confirm I have good instincts, because I really did find this work to be extraordinary and fascinating.
Well, depending on how broad your definition of life is, viruses have the most stripped down genomes of all. In the smallest viruses, with genomes of just under 10kb in length, nearly every basepair is dedicated to either infection or replication. In fact, they are often so compact that open reading frames are interleaved, in order to provide more functionality without increasing size.
Scientists often refer to viruses as "obligate", in order to sidestep the question of what is life, as most have no interest in the topics which occupy philosophers. In any case, they are non-cell based, for whatever that is worth. I imagine in a non-hostile environment, even the infection functional would be shed, and you would be left with just replication, which is the fundamental component beyond which no further reduction in complexity can be made.
> I imagine in a non-hostile environment, even the infection functional would be shed, and you would be left with just replication
A virus replicates by infecting another cell and taking over its actual replication infrastructure, so getting rid of infection gets rid of replication too.
> viruses have the most stripped down genomes of all
Giant viruses can have over 1M basepairs, substantially larger than a bacteria such as Mycoplasma genitalium, with substantial functionality (pretty much everything except the ribosome in at least some of them: https://www.virology.ws/2018/03/08/only-the-ribosome-is-lack...)
> I recalled cutting edge experiments using synthetic cells to create artificial life (a worm perhaps with a relatively simple DNA, maybe even a modified DNA further simplifying the genome to the furthest extent possible still resulting in life)
A worm seems super-complex for something like that. I'd guess they'd actually use a bacterium.
I thought some work on it made HN, but can't seem to find the article, about a research group that was continuing to strip things out and then test viability.
I wish I could be more helpful, but this was during a period I was distance running everyday for years without much concern on what I was consuming so long as I found it interesting.
In terms of popular podcasts maybe I could say Lex Friedman, but then I might search for one of his guests, or specific topics I wanted to learn more about, on YouTube and look for lectures or panel discussions in the results that looked like they might be high quality.
Not sure if you’re doing it a service or disservice calling it chaos engineering: at best it’s related. If you want to figure out what a complex system does where you have no ability to “see”, the only tool you have most of the time is to knock out individual components and see what happens.
As you might guess this is generally a blunt tool which can help you get to the first 30% of the understanding of the system but minimal extra data after that. The majority of genes discovered in this study would either be already known players in those pathways or unknown genes that would already have been guessed to play a role.
Until one of these massive screens tells us what the major vault protein complex does they should all be honest about what they are which Imo is just a minor addition.
It says that some genes result in the same outcome when knocked out as other genes, and identifies novel genes that putatively participate in the same pathways as others. This helps get at the potential function of genes without known functions.
So, people do this a lot and frequently make mistakes. For example, when you knock out a gene, you also damage any overlapping genes (yes, genes can overlap). most studies don't pay attention to the damage they do to overlapping genes.
The underlying physical model for how gene products interact to make phenotypes ends up being so hopelessly complex and latent that most conclusions in this area end up being "sufficient, but not necessary" instead of "necessary, but not sufficient"
> It says that some genes result in the same outcome when knocked out as other genes
a very useful map to make, but I don't see that this contradicts my comment - both genes in this case are dependencies to the outcome, without either of them, the outcome fails
this does not sound to me like we know the "function" of these genes, only that they're nessary for each phenotype
not to knock the research, just trying to make sense of what they're really mapping, in my own language of computer code (i suppose "function" has a different connotation in genetics)
> this does not sound to me like we know the "function" of these genes, only that they're nessary for each phenotype
But they're not even measuring the phenotype. They're using the transciptional signature as a substitute for phenotype/cell function (i.e. the bag of RNA model). This is a poor substitute if you try to apply this to practical applications such as cell engineering. Let's say I perturb a cell to match it's transcriptional signature to that of a neuron. Does that make it a neuron? Not if it doesn't function like a neuron.
I think this is a really important point. With a few exceptions (like the neat implied aneuploidy assay), they haven't measured an outcome or phenotype for the genes. They have measured the impact on transcription (well, mRNA levels, via transcription or some other effect). That is an extremely useful dataset, but it's not enough to say what the phenotypic effect of knocking out any given gene is, much less what the actual mechanistic function of the gene product is.
It's also important to note that there are loads of genes whose effects are not mediated by changes in mRNA levels. If you knock out Arp2, a cell can't move properly, because Arp2 is involved in assembling cytoskeletal structures needed to do that, but you probably won't be able to tell that by looking at the cell's mRNA.
> "this outcome fails when this byte of data is missing"
The outcomes here are not failures, they are measurable phenotypic differences, which they use to group genes into phenotypic outcomes. The typical "knockout -> failure to perform a function" is not what's being measured here.
It's kind of a dead end I feel. It can turn out that everything is used for everything. It's not designed so there is probably no clean decomposition nor is it necessary.
Turn off the genes that would make my child below 6'5 in adulthood, turn off the genes that wouldn't make him naturally muscular, turn off the genes that would give him an average intelligence, etc.
This doesn't really pass the sniff test; if there were any easy one-gene changes that resulted in unambiguously better results, it's highly likely that we'd all have them already.
Well, these aren't unambiguously "better results", not in a biological sense. Just better in terms of what our culture values, and not necessarily better survival and reproduction.
BCRA mutations: "It’s estimated that 55 – 65% of women with the BRCA1 mutation will develop breast cancer before age 70" [0]
ApoE4: "individuals with two versions of the ApoE 4 gene have a 50% chance of developing Alzheimer’s" [1]
HTT: The Huntingtons gene. If you have the wrong version of this gene, you will get Huntington' disease.
All that to say that, in these cases, sure _most_ people have positive traits from these genes, not all do and we could theoretically use CRISPR to change people with the negative versions of these genes to the positive versions of these genes.
It's embarrassing when self-appointed "bioethicists" do nothing but handwring about distant transformative technology - stuff that could confer such incredible benefits on humanity and that we are light-years from having.
It's so easy for them to posture as virtuous protectors of humanity, sowing FUD about things that are so far from existing that nobody even understands how they might work or what capabilities they might have or not have. So easy, when the benefits are nowhere near close enough to factor into anyone's near-mode thinking.
I'm cheering for the scientists. Give us the choice to have brilliant, strong, healthy children. Then laugh at how little anyone cares about the so-called "ethics", facile cloud-talk that never considered real people facing real decisions.
The "ethicists" had their chance to intelligently weigh the pros and cons and give real advice based in prudence, empathy, and a love for human flourishing. They chose to self-indulgently chinstroke about dystopia on the taxpayer dime.
$10M+ now and then in 20 years figure out that changing that gene sequence creates a dramatically higher likelihood for Alzheimer's or infertility. I personally am not against the modification of humans in practice, I think it's an interesting opportunity for humans to take control over the evolutionary process and optimize ourselves for the world we live in now (and future ones), e.g., what if we could minimize the amount of sleep needed, or our tolerance to heat so climate effects were somewhat mitigated, or so our bodies were less affected with long periods in space? The tie-in of capitalism to it creates a weird dynamic but I'm not sure I see an obvious better solution unless we want to outlaw everything except for government approved research and let the ideas be at the whims of whatever bureaucrat or political appointee happens to be in charge at the time.
>$10M+ now and then in 20 years figure out that changing that gene sequence creates a dramatically higher likelihood for Alzheimer's or infertility
Not only that, but any genes that are not selected against (e.g. your Alzheimer’s example, which occurs after reproductive age) are now permanently circulating in the population, with absolutely zero (humane) way of ever removing them.
In the short term, a lot of gene therapy focuses on the set of diseases where the repair really does seem to be "change this one base" and there are no side effects due to complex, unknown gene interactions or other unexpected phenomena. m And nearly all of it (I haven't looked at the details in a while) is just treatment (IE, injections of a gene therapy in a post-birth individual), rather than preventive by germline manipulation, which IMHO
It's likely that for the time being we'll only use gene therapy for things which are recognized as devastating diseases and the treatment is extremely reliable; germline modification for enhanced attributes in typical individuals is still a fairly out-there concept that would probably get panned in the media.
I have seen Gattaca and don't find human genetic engineering to be horrific. It's cool and exciting that people will be able to have better children than they would have otherwise.
> I have seen Gattaca and don't find human genetic engineering to be horrific. It's cool and exciting that people will be able to have better children than they would have otherwise.
Did you even pay attention to the movie? The horrific aspect was the human genetic engineering led 1) to the the unengineered to be turned into an underclass that was blatantly and unfairly discriminated against, and 2) that same discrimination would be turned against the engineered if they had an accident that caused them to fall short of the expected perfection.
I have little doubt that the reality of genetic engineering (that was is effective as that depicted) would rhyme with that movie. It's also nearly certain that any such technology would not be distributed in an egalitarian way, so the sentiment should be more like "it's cool and exciting that [well off] people will be able to have [genetically superior] children than [the plebs]."
I understood the intention of the movie, but I'm also in the camp that thinks engineering ourselves is the way forward, and a given. Genetic and silicon advances have a real potential to make us super/transhuman and immortal.
> It's also nearly certain that any such technology would not be distributed in an egalitarian way
This argument could have been made about early computers as well. But time and its economies of scale come into play. We can't limit our species based on shortsighted fairness, there's a longer view to take.
> I understood the intention of the movie, but I'm also in the camp that thinks engineering ourselves is the way forward, and a given. Genetic and silicon advances have a real potential to make us super/transhuman and immortal.
Techno-optimism hasn't really panned out as promised.
And I really doubt it will be "us." If trans-humanism pans out (though I suspect it's bunk), we'll be the Homo erectus populations to their Homo sapiens.
> This argument could have been made about early computers as well. But time and its economies of scale come into play. We can't limit our species based on shortsighted fairness, there's a longer view to take.
Capital got automation from computers, the plebs got distraction boxes that push ads.
Hrm. So what? How is that different from the uneven distribution of money/wealth/power/medicine/upbringing/private schools/university/"connectedness" today?
This will be just another currency. And there will be some flops, busts, unintended consequences. Just like with anything else.
> Hrm. So what? How is that different from the uneven distribution of money/wealth/power/medicine/upbringing/private schools/university/"connectedness" today?
Vague resemblance does not an equivalency make.
Right now, there may be an unequal distribution of wealth, etc.; but the wealthy by and large aren't actually better models of human. Genetic engineering has the potential to ossify that (both morally via "meritocracy" and physically) with biological superiority. It stops being Homo sapiens vs Homo sapiens, and becomes something more like Homo erectus vs Homo sapiens.
We have underclasses based on genetics now. Let's not pretend mate selection on Earth in 2022 is a meritocracy. It's mostly just racism, a slower form of deliberate genetic engineering of one's offspring.
If anything, this would increase the amount of opportunity in the world, as then your child could have whatever traits you (or your culture) deems superior. It would perhaps eliminate racism by rendering traditional race markers completely obsolete.
> It's cool and exciting that people will be able to have better children than they would have otherwise.
You know, I used to be wary of eugenics, but when you put in that light, yeah, I'm kinda tired of putting up with everyone's crap kids. If you could make them be quiet, sit still, and do as told, that would be a fantastic achievement! Oh, and maybe make them smarter, too. Ever try to actually talk to one (the newer ones are really pretty stupid)?
People will balk at this, which is funny in contrast to how much human energy is poured into ego, signaling and mate selection as proxies for perpetuating favorable genetics.
- this is performing an irreversible procedure on a person without their consent. Justifiable for a disease but less so for vanity traits.
- genes are pleiotropic and modifying multiple genes could have unexpected effects. An easy example is that taller people have a higher risk of cancer. Barring some great advance in predicting the effects of genetic changes, this will remain a problem.
- it has a homogenising effect, optimising for traits a culture finds appealing and that can be easily measured. You will eventually get a eugenic army that all look like swimwear models and score highly on IQ tests.
At the same time, I think that modifying the organism is the only way to get rid of diseases like heart disease and cancer.
Given that people already travel for IVF to cheaper locations abroad (Ukraine was one of them until the war), it is going to be more like $100K+ in the future.
Medicine is practised outside the US as well, including high tech medicine. And for a much more reasonable price tag.
I think so. Alexander Graham Bell, Helen Keller, Winston Churchill, Plato and there are countless more. Just because we didn't have the technology to implement it humanely in the past doesn't mean it can't be done in the future.
In fact. I would consider a society, that has the capability to noninvasively eliminate for example sickle cell anemia or Huntingtons and doesn't because some people one hundred years ago did horrible things, barbaric.
Helen Keller defended a doctor's 1915 non-intervention in the case of a child that was not capable of "the possibilities of happiness, intelligence, and power that give life its sanctity, and they are absent in the case of a poor, misshapen, paralyzed creature", and then later supported adoption of disabled children, such as the a case of an infant with tumor-induced blindness saying "blindness is not the greatest evil, it is only a physical handicap. that is life. the annals of progress show that much of humanity's finest work has been wrought by persons with with a severe handicap, that she may be spared to help open the eyes of ignorance"
I hope that she can help you to open your eyes as well.
Sorry if i hit a nerve, but obviously if i wasn't "arguing in good faith" why would i have so painstakenly avoided mentioning hitler for like 10 comments?
conversely, everything i've said has been a direct analysis of what you've said.
you referenced helen keller's non-intervention in an instance of infant mortality as evidence that she supported eugenics.
It is that erroneous conflation that introduced euthanasia for discussion. (also i would say that technically intentional non-intervention is not exactly killing, so thats not even what i was implying)
I provided evidence that keller, contrarily, saw value in genetic defects.
If you have another reason for believing that helen keller was a supporter of eugenics, you have failed to provide it.
Churchill had some virtues of leadership, but he was an abominable racist and colonialist, even by the standards of his time.
>"Churchill is on record as praising “Aryan stock” and insisting it was right for “a stronger race, a higher-grade race” to take the place of indigenous peoples. He reportedly did not think “black people were as capable or as efficient as white people”. In 1911, Churchill banned interracial boxing matches so white fighters would not be seen losing to black ones. He insisted that Britain and the US shared “Anglo-Saxon superiority”. He described anticolonial campaigners as “savages armed with ideas”.
Even his contemporaries found his views on race shocking. In the context of Churchill’s hard line against providing famine relief to Bengal, the colonial secretary, Leo Amery, remarked: “On the subject of India, Winston is not quite sane … I didn’t see much difference between his outlook and Hitler’s.”"
>"He referred to Palestinians as "barbaric hordes who ate little but camel dung." When quashing insurgents in Sudan in the earlier days of his imperial career, Churchill boasted of killing three "savages." Contemplating restive populations in northwest Asia, he infamously lamented the "squeamishness" of his colleagues, who were not in "favor of using poisoned gas against uncivilized tribes.""
Also, Alexander Graham Bell supported eugenics because he was worried that deaf people procreating created human and social divisions.
Any modern medicine gene modification program would create the exact human and social divisions he sought to avoid... Unless you are somehow able to suddenly do gene modification universally throughout the entire planet..
so he doesn't really support your position either.
it has taken us 2 years to give 60% of the planet a $5 shot despite having 2x world population of shots available.
"capability" should not be confused with "universal accessibility"
and "some people one hundred years ago did horrible things" is a fictional strawman. There are real and present reasons why practically eugenics will lead to stratified and unjust conditions for humanity in practice.
Not just wealthy couples. Most parents want their children to be heterosexual and bear them grandchildren. Having successfully procreated, they are more likely to have natalist genes and inclinations.
Why would they need to be heterosexual to successfully procreate.
If the future is designer babies/baby incubation pods. It doesn't matter what sexual orientation the kid is. The parent would have a designer baby, then why would the child eventually not have a designer baby as well?
You will always need an Egg or Sperm depending on which is missing in the relationship. Also, one of the parents would not have genetic code in the offspring since you can't splice two eggs/sperm together as far as I know.
We already selectively breed for natalists and look how that turns out. It actually makes an argument that genetic predisposition to natalism is currently rather weak and it is mostly memetic predisposition in practice.
The demographic trends we're seeing now (declining birth rates for affluent people in the developed world) could be due to our natalism level being mis-calibrated for our current circumstances.
It's good evolutionary strategy to reduce the birth rate to compensate for successful reproduction (i.e. more pregnancies that result in offspring that reach child-rearing age themselves), as otherwise populations will explode beyond the carrying capacity of the environment (or pay more of the fixed resource costs of raising children than is necessary).
What we see now, however, is that in situations where both pre-adult and maternal mortality rates are exceedingly low (such as has been the case for the last ~4-5 generations in the developed world), the strategy undershoots the replacement rate.
Yea human culture will find a new level of 'basic' combined with an inferred and understood if not blatantly codified position of social superiority... Sounds more like twilight zone than social progress to me.
This already is what happens with sexual selection. That is the purpose of sexual selection from the point of view of genetic integrity. People are designing their babies every time they have sex. There's huge amounts of money involved there too. And like how do I know the geneticist isn't slipping his own in-born DNA in my place, like that fertility doctor did or that criminal inseminator who broke into poorly-guarded sperm banks and substituted his sperm for that of astronauts with Nobel Prizes? Had like 660 kids that way.
Like there good reasons to do this are, OK genetic malformation, ie osteogenesis imperfecta, that's a SNIP, totally. Or you were raped, want to punish that rapist by making the egg have even less genes in common with him than it would ordinarily, making his rape counterproductive genetically. Another good reason, defensible. Or you want to protect rare alleles, considering weaknesses to be strengths, this is like a left-handed man marrying a left-handed woman, or freckles (I think something about freckles make children hate them til they're 19, rare exceptions, not even the iconic Lindsay Lohan accepted her own freckles). Preserving their identity and diversity. Don't pass laws as much as using judgment to decide, individually and collectively, what the future will be. Talk and think instead of setting up magic words for lawyers to copy-paste--an isomorphic problem to genetic design.
No fucking blind copy-paste, don't use copy-paste with your copy paste to copy-paste yourself.
I'm not a biologist. Still, do genes really have "functions"? Guess some have, such as producing an enzyme to break down a particular kind of suger. But that this would be the case for every human gene sounds unlikely to me.
For your last sentence: that's right, to get any press attention today you have to basically say you've solved all cancer or done something pan-genomic, in a way that massively overstates the importance of the specific result and its impact on health care delivery.
That said, Jonathan Weissman is a great guy who has pushed the field forward and the techniques they are using really are powerful.
A gene is a DNA (or RNA) sequence that codes for some protein or RNA product. There are other functional sequences in DNA but they're not called genes.
All genes have a function. This paper helps us understand genes of unknown function (which is a LOT of them).
Some of these functions are not intuitive: maybe they regulate the function of another protein, maybe they only function in the context of a particular stressor, etc. You can think of nearly unlimited scenarios to apply and you start to understand the complexity of understanding how a gene functions.
"All genes have a function" <- this is a generalist statement that is wrong in its specifics, but also raises semantic question of what "function" actually means.
A lot of studies have shown that literally taking up genomic space is a function. So by definition, all genes have a function. Non-genes have functions. The topological organization of genomes is a function.
This is a bit of a semantic argument, but gene function is a fairly nebulous term. The essence of what I am saying is that there may be proteins that currently have no actual function, aren't under functional selection, yet are duplicated, transcribed, and expressed (not just pseudogenes).
Function is a rabbit hole. Biologists get in big arguments about the semantics of this all the time (http://cryptogenomicon.org/encode-says-what.html). I don't really care. I care about the minimal set of necessary proteins for a model organism to exist and reproduce in a media-rich environment. And, whether there are actually subsets of mutually compensatory groups of proteins instead of a single minimal set.
Protein function is one of those things that, at first, seems really simple to define, but the further you go down the rabbit hole, the more complicated it gets, until it's fractally complex and you realize that not only does the exception prove the rule, it's all exceptions.
I think the semantics are largely dependent on the level of abstraction you are reasoning at. Here I use abstraction loosely to refer to conceptual or physical scale / granularity / resolution. For example, we are (probably) both using the term “gene” as shorthand for the concept of “gene or gene product(s)”. Likewise the term function can refer to phenomena occurring on molecular, cellular, etc length scales, or more amorphous groupings of phenomena such as “flux” through a biological pathway.
But if you really drill down into the nitty gritty, then the “function” of a “gene” is its complete set of state altering / modifying relationships with other bio entities. In this sense, all bio entities have functions because they all have functional relationships with other bio entities.
So yes, all genes, pseudo genes, isoforms, etc have a “function” even if it is redundant, taking up space, or just soaking up some of the pool of tRNA.
Also the minimal genome stuff is pretty fascinating! One of the best research questions I’ve ever heard was, “what are the essential genes of unknown function doing?”
The problem is that proteins do some things passively unrelated to (say) their enzymatic ability. Is that secondary functionality a function? What if it's binding a molecule, but then releasing it before the catalysis occurs (wasted effort).
I mean, I know a person in grad school who worked on finding the function of a protein for a long time. It was given by a collaborator and had high sequence similarity to a known enzyme in a related species. She tried every possible functionality test to see if it was a protease, or any of a hundred other enzymatic reactions. Eventually it turned out the collaborator had mistakenly given them an alanine-scanned protein with the necessary functional residues replaced, so she never detected any activity because there wasn't any. Does that mean the protein had no "function"? It was binding water molecules, even plausible substrate, but just never helping a transition state form. Even if you replaced the working version of the protein with the broken version in an organism, if it wasn't a completely necessary protein, it would continue to reside in the genome with no function for some time until (perhaps) neutral mutation due to lack of functional selection caused the protein to be non-expressed and it starts to rot away into a pseudogene.
The main problem with your research question is that it still hasn't been completely resolved- there are proteins remaining which are necessary, but their functions are unknown.
For the benefit of readers unfamiliar with the field: this is wrong.
You can show this experimentally. Construct a gene that produces a non-human protein and introduce this to a human cell/genome. That gene would not have a “function” but still exists in the genome. This is actually happening all the time. Some viruses integrate their genomes when infecting cells. Viral integration is one of the factors that shapes genome evolution.
Even an exogenously expressed non-human protein still has a "function". even if you force it to be expressed. We may not understand its effects, but it is certainly doing something. Even just taking up a space is a biological function, which an exogenously expressed protein is doing. The same applies to genes that do not get translated to protein, by definition of "doing something" they have a function.
Your viral integration example is actually a perfect example of one where all genes indeed do a have a function, but they are not readily apparent to us. Genes that control latency may not expressed until specific conditions, and that is their function, to control expression. Some genes control integration.
I spent 5 years of my life doing my PhD studying viral replication, and the "unknown function of viral genes" was a constant topic of discussion, but we all agreed, they have a function.
You're conflating "consequences and effects" with "function". The former is things that happen due to the physics, the latter is about intent or utility.
"Taking up space in the genome is a function" is a great example of this. While I'm sure you can find examples of "spacers that when deleted are fatal", the fitness effect of protein-coding regions that contain no utility is still an area of research. To me, functionality requires selection, although that's probably not necessary or sufficient!
A gene is essentially defined as having a gene product, either an RNA or protein. So yes, all genes are functional.
Genes that aren't translated into protein sequences (noncoding genes) can create structural RNAs as with Ribosomes, microRNAs that have regulatory functions, etc.
There are lots of non-gene genetic elements that do things, though. Many are involved in gene regulation, affecting the transcription rates of nearby loci through a variety of mechanisms. There are also vast swaths of inactivated transposons, retroviruses, and other repetitive genetic sequence.
His explanation in GEB of how genes, DNA, and RNA function was probably the clearest one I’ve ever read. It’s been a while and I forgot the details, but reading his analogies made everything click at the time. Maybe it’s time to re-read GEB.
Never underestimate MIT’s ability to oversell itself.
(EDIT: see below, this is directed at the press release, which I perceive to be overstating the achievements presented in the paper, not the quality of the paper itself)
Sure. A genome-wide perturb-seq experiment is a huge (and expensive) technical accomplishment, but the authors did not "map every human gene to its function". (Nor did they claim to, it's press release hyperbole.)
One, there's ~20k protein-coding genes in the human genome, and they screened ~10k, analyzing about 2k (fig 2a).
Two, all the functional annotation is based off transcription profiles. They essentially looked for clusters of genes with correlated expression, and assigned function based on genes with previous annotations (fig 2d, S4).
It's a good resource, but there's a lot more molecular work to be done to validate the function of these genes.
> Two, all the functional annotation is based off transcription profiles. They essentially looked for clusters of genes with correlated expression, and assigned function based on genes with previous annotations
This is an important point, because if you've ever worked with single cell data you'll know that the transcriptional profile is extremely noisy and your transcriptional profile to cell type map has many researcher degrees of freedom. I heard a story about a paper early in the single cell work that started with 53 cell types and after review ended up with 37 cell types. Are those true cell types? Did the experimenters validate that those cell types all performed different functions? Well, of course not. That's way too much work.
Then add on technological biases, which make mapping between technologies difficult. I say this because they used a new sequencing technology that appears to have homopolymer bias (https://twitter.com/lpachter/status/1533875723995185153), which will bias the gene quantification.
I believe they used Illumina for the results presented in the main text and then re-sequenced with Ultima and replicated a subset of the analyses (fig s13). The Ultima proof-of-concept didn't appear to be relevant to the main study/conclusions
Thanks. I'll admit the first comment didn't contribute much and apologize for not including my thoughts from the follow-up. I just don't care for these sorts of hyperbolic press releases.
It's part of the game, the big players are as good at sales as they are at science, but I've never been a fan of it.
Useful book that any starting professor should understand, even if they don't want to admit they are playing a game. science-by-press-release is a technique to master if you want to both maximize the impact of your work, and get tenure.
An article from MIT:TR in the early 2000s always stuck with me as an example of how intricate and interconnected genes really are. Researchers using fruit flies found one born with white eyes, and they narrowed the mutation to a specific gene. They were able to modify it in vivo to reliably produce white eyes, but the change had an unexpected second outcome, which was that white-eyed males would only attempt to mate with other males.
Years later, there was a group who cited that fruit fly paper when they proposed the same methodology to control mosquito populations, but I'm not sure if they ever recreated the male preference. The mosquito gene editing did pan out, but the method is different in that it doesn't allow females to survive, while males will go on to mate with other non-edited populations and spread the female-killing gene.
> New CRISPR-based map ties every human gene to its function
The title would be so much better if s/its/a/.
There is no way we'll have an operative map from "all" genes to function anytime soon. Sometimes hundreds of genes work together, interact with other microbiomes we contain or our environment to produce what might seem to be a "simple" quality like height.
Of course I was aware of the Human Genome Project and mapping DNA in general. I was also aware that figuring out all the proteins in a cell and what they do is a whole other problem.
I didn't realize they'd made this much progress. It's not complete obviously but being able to figure out gene expression is a massive step forward. The ability to switch off genes (this is where the CRISPR editing comes in) and seeing what changes is just astounding (there was an example of chromosome segregation).
It's known that certain proteins mediate certain processes where the presence of that protein or the absence of it can lead to a condition or disease. The potential impact here for treating genetic disorders I think cannot be overstated.
Between this an the technology behind mRNA vaccines, I really wonder if the 21st century will lead to the effective elimination of many diseases.
The 21st centry may lead to the definitive elimination of humans, of desease i can only imagine their numbers going up, not down.
"that's where CRISPR editing comes in". To edit the DNA of congeniality diseased people? I doubt the business will focus primarily on this category of patients, unfortunately. but I still hope the worst will be avoided when/if using this tech kind of tech.
If they approved the sort of recent treatment they approved, for that scale of deployment (billions of individuals within a year) it wouldn't surprise me they will approve anything, given enough lobbying. No worry to have with FDA becoming an obstacle.
We've being doing genetics for decades. Molecular biology without mutant studies wouldn't exist. It's the foundation of the field.
All this is is a difference in scale. But it is a very crude tool; really understanding gene function involves studying it in relevant contexts. Looking at cells in tissue culture can give you some ideas or hints about how it functions, but the critical insight might require certain cell types or gene regulatory environments.
What data like these do is inform hypothesis generation and refine the interpretation of genomic data. It is important work, but does not replace doing actual biology.
What you're talking about is essentially saying we just invented molecular biology. Which is obviously not the case.
> The ability to switch off genes (this is where the CRISPR editing comes in)
i think your wording is ambiguous : CRISPR can edit the sequence, but "switch off" has an association with methylation of items in the sequence, orthogonal to the sequence itself.
> I didn't realize they'd made this much progress. It's not complete obviously but being able to figure out gene expression is a massive step forward. The ability to switch off genes (this is where the CRISPR editing comes in) and seeing what changes is just astounding (there was an example of chromosome segregation).
The protein folding problem was solved earlier this year. You can expect a lot more coming in this vein... interesting times.
the protein structure prediction problem was solved, not protein folding. Different field. Also, it wasn't really much "solved", so much as prediction got as good as the metric used to compare the predictions to reality, so we can't really say for sure whether the predictions are better than golden labels or not.
Would love to dive in looking into the data, but I don't understand it. Could someone share how to "read" the data file both from non-technical and technical perspective. Right now not sure how to interpret the data and I also do not have a biology/gene background.
Do you have a direct link to the data you mention? Something more specific than a general “help! I am lost!” would go a long way. I dont really see it in the article.
So, like... I'm pretty ignorant on biology, genetics, gene editing, and such. But... could one alter their genetics for, say, poor eyesight and eventually have better eyes? I mean, I imagine that there is a shelf life for cells already in the body and they don't just go away right away. But they are also constantly being replaced, right? So... can I just... rewrite my genome to replace my eye cells with "better" eye cells?
And--hypothetically speaking, for sure, certainly, I assure you--who might be capable of such a thing, and what might they like as remuneration?
This is akin to taking an automobile apart piece-by-piece, not to determine its function but just to figure out if the automobile fails substantially by removing say a taillight, dome light, dash light, or wiper delay fuse.
The car's not gonna fail for each piece, it'll take forever to determine what pieces are absolutely necessary, and it doesn't tell ya shit about what the pieces functions are.
Neat, but ultimately inefficient and exceedingly limited in necessary detail to make the claim in the article's title.
This is cool, so in laymans terms - you can disable the bad "genes" and just express/enable/push the "good ones."
So how do we define what is a bad gene? If we use crispr, can we turn it off on an actual, live, aging human being or only before they're "born" or such?
Is there a good resource/book that gets someone from zero to a basic biological understanding and background of this and above? I really have no idea, nor do I know who to ask. :(
Q: How can the human genome be "mapped" without sampling and sequencing genetic material from millions of individuals? Is it a sequencing of one individual?
Anyone know where the state of the art is happening in applied 'designer baby' tech? I feel like there hasn't been much news since the arrest of the scientist who claimed to be the first to change the genes of live kids but I'm shocked there aren't some people out there pushing the envelope on this (and/or commercializing it).
Very frightening … good to know relativity but then we have nuclear bomb. Not sure the negative side to this. Just hope the good side out win any bad use of information about all of us.
One model of how science progresses is a few pioneers breaking new ground and then everyone else rushing in to pick up low hanging fruit in the new field.
Looks like Weissman lab has consistently been breaking new ground over and over again. Extremely impressive and very few labs in the world have such a track record.
Maybe it's because I'm not in that industry, but I was looking for a graphical gizmo that I could click on a gene and see functions but they mean "map" in its functional programming term
Although clicking through the first link does say "Interactive Website under construction..." so maybe this was just submitted too early or something