'Obelisks': New class of life has been found in human digestive system
476 points by unkeen 6 months ago | 134 comments- ababaian 6 months agoCool :) I'm a co-author on this. AMA.
This is now a peer-reviewed paper, published last month in Cell [https://www.cell.com/cell/fulltext/S0092-8674(24)01091-2].
Obelisks are part of a larger research program we're developing at the University of Toronto + collaborators, see also: Virus-Viroid Hybrids paper [https://www.nature.com/articles/s41467-023-38301-2] and the Zeta-Elements [https://www.nature.com/articles/s41586-021-04332-2].
Computational biology is driving a revolutionary expansion of our understanding of Earth's biodiversity. I believe Zeta-elements, Ambiviruses, and Obelisks are just the beginning. If you're interested, our "Laboratory for RNA-Based Lifeforms" (University of Toronto) is hiring passionate developers/post-docs/graduate students [https://www.rnalab.ca].
Edit: OK going to call it for now. I'll check in later today if there's any outstanding questions.
- marojejian 6 months ago>Obelisks form their own phylogenetic group without detectable similarity to known biological agents.
ababaian, does this truly mean no similarity to any other sequences, even virus/viroid?
That seems very exciting, since my understanding is that we see a lot of conservation within the known branches of life, and don't discover new ones often!
Though perhaps it's more common to find totally novel virus/viroids? How often do we find truly novel biological agents at the sequence level?
- ababaian 6 months agoTo the limits of where our understanding of how "entities" are connected to one another (homology). Yes, there's nothing like them. You could make an argument they are "viroid-like" and there's a deeper evolutionary connection between viroids/viruses/plasmids, but the information theory to formally establish such a connection is not sufficiently developed. It's a worthy scientific problem!
Is it common to find new viruses/viroids/biological agents? Well it certainly is starting to feel that way to me.
- DoctorOetker 6 months ago> You could make an argument they are "viroid-like" and there's a deeper evolutionary connection between viroids/viruses/plasmids, but the information theory to formally establish such a connection is not sufficiently developed.
The way it is phrased, insufficiently developed information theory is rather surprising. Did you mean to write that not enough genome data has been collected to formally establish a link, or are you actually stating that we have all the data but as a species have not sufficiently developed the mathematical subdiscipline of probability, information theory ?
I could follow the first, but the latter?
EDIT: I now believe you meant neither but more something along the lines of: we probably have plenty of data, and usual information theory should suffice, but we simply havent exhaustively applied the tools to collate the information and make the implicitly available data more explicitly manifest.
- DoctorOetker 6 months ago
- ababaian 6 months ago
- Traubenfuchs 6 months agoHow do obelisks fit in with commonly known forms of life?
We all understand cells/bacteria and their interaction with viruses: Viruses infect cells and make them into virus factories…
What do obelisks do? Are they integrated / read by DNA machinery/organells into cells that then produce more obelisks?
What‘s their life cycle?
How are they different from alread known viroids?
- ababaian 6 months agoLargely we don't know. That's what makes them cool. This opens up a fairly large debate about where you draw the boundary of "life", "organisms", and "entities". It's a fun debate to have over a few beers since there is no objective correct answer.
My view is that Obelisks are more like Viruses or Viroids, or some kind of mobile genetic element. The key detail is that they appear to be strictly RNA elements (they don't have a DNA counterpart). So they're most likely using host RNA transcription machinery to make more copies of themselves, this is what viroids and satellite viruses like "Hepatitis Delta Virus" do.
What do they do? Well that's the right question. My guess is the kinds of things that bacteriophages do, Obelisks do too. Exploit cells to make more copies of themselves as selfish replicators.
- brokensegue 6 months agoI'm unclear why we are calling these a new class of life rather than just a new kind of virus. Their shape?
- bicx 6 months agoIs it possible that these may cause diseases that previously went unexplained?
- brokensegue 6 months ago
- ababaian 6 months ago
- trebligdivad 6 months agoHow do you know that they aren't waste/intermediate products of some other cell, as opposed to being something that reproduces?
- ababaian 6 months agoHow would they exist/maintain themselves without it's DNA counterpart?
- trebligdivad 6 months agoah, have you shown that there is no matching DNA in either the host or any of the bacteria that the host has?
- trebligdivad 6 months ago
- ababaian 6 months ago
- stainablesteel 6 months agoin simple terms, it seems these are just rod-shaped RNA plasmids that encode a couple proteins and exist without any kind of membrane or coating, does that seem right? is it that elucidated?
- ababaian 6 months agoSure, in a redutionistic sense. In the same light "Hepatitis Delta Virus" is an RNA plasmid, yet it causes liver cancer in humans. I err on the side that the simplicity of the genetic system should not deceive us into thinking it's trivial. The next 12 months of discoveries is what makes this so exciting.
- ababaian 6 months ago
- yawnxyz 6 months agoAre there more of these undetected/undetectable entities out there?
E.g. are there more "life" like obelisks and similar out there in genome samples?
- throwup238 6 months agoProbably. There’s a whole category of organisms that can’t cultured in a lab or effectively studied called microbial dark matter: https://en.m.wikipedia.org/wiki/Microbial_dark_matter
These kinds of DNA and RNA studies are the only ones that can realistically pick up evidence of these organisms outside of an extraordinarily lucky electron microscope slide.
- hinkley 6 months agoThere’s a biological detective story out there about someone figuring out how to culture the black soot near distilleries. This isn’t the version I read but it’s the same microbe:
https://www.wanderingspiritsglobal.com/whisky-fungus-baudoin...
It doesn’t culture in agar. Unless you add alcohol to the Petri dish and then it does. Things like this are why I still have my fingers crossed that we will have one or more helicobacter pylori (the bacteria that causes 90% of ulcers) moments for intestinal ailments. These obelisks may turn out to be one of them, and understanding could lead to better treatments and prevention.
- ababaian 6 months agoIf you want the real mind-trip, try to think about how little we know about detecting "life" that is not based on nucleic acids.
- hinkley 6 months ago
- ceejayoz 6 months agoProbably. This is a good illustration of just how hard it may be to conclusively find life on other worlds; we don’t even know what exists here on Earth yet.
- light_hue_1 6 months agoThe reason these things are so hard to find on Earth is because it's teeming with other life. If it wasn't and these things were isolated on Mars it would be relatively easy.
- aaroninsf 6 months agoThis is basically the core plot device (and foregrounded assertion, for its ramifications) of a recent scifi book, which for spoiler reasons I rot13:
Xvz Fgnayrl Ebovafba'f Nheben
- light_hue_1 6 months ago
- ababaian 6 months agoI should certainly hope so.
- casenmgreen 6 months agoDon't know. They're undetected =-)
- throwup238 6 months ago
- bryan0 6 months agoHey this looks pretty amazing, congrats on the research. The article doesn’t seem to explain much though and the papers are pay-walled(?) so my question is how were these discovered only now and not noticed before?
- ababaian 6 months agoThere are preprints for all the papers [https://www.biorxiv.org/content/10.1101/2024.01.20.576352v1], they're operationally the same.
How were they not noticed before? Well that's how science works. Someone eventually has to be the one to notice something is going on right?
I think it's a common fallacy that we, as a species, are not ignorant to the complexity of Nature. The hardest part is to see it.
- layer8 6 months agoI would expect that you still need to have some substantial expertise to have any idea of what you are doing, and to know that what you are doing makes sense and also hasn’t been done yet. Someone unfamiliar with molecular biology would likely need to invest some time to get up to speed.
- DoctorOetker 6 months agoConsider the existence of forensic genetic polymer tagging sprays etc.
For example: https://www.selectadna.co.uk/dna-tagging-spray
One could easily fathom not just overt authorities but also covert authorities wishing to use similar technology.
Clearly an intelligence agency doesn't want the lower level police leaking detections of higher importance, so best to differentiate say DNA for cops and RNA for intelligence services, so that the pragmatic tools and workflows of police won't result in uncleared personnel figuring out things they aren't cleared for.
That wouldn't explain quasi biological statistics as opposed to white noise random sequences, which would suffice for tracking, but also would blow the cover as a man-made genome...
I.e. if RNA sprays had been reserved by some power bloc for intelligence service purposes (DNA sprays for usual law enforcement), then there is a clear incentive to have the secret pseudorandom sequences at least mimic plausible biology sequences, an adversary bloc detecting such a tracker might believe the sequence to be of biological origin, and intelligence associated academics would publish it: revealing both the detection by an adversary bloc and the academic's employment by intelligence circles...
- layer8 6 months ago
- ababaian 6 months ago
- andrewflnr 6 months agoDo you have any clear idea yet what the proteins do that the obelisks code for?
- ababaian 6 months agoNo clear ideas. But I would not be surprised if a dozen labs have jumped at the chance of trying to figure it out. So wait a year and ask again.
Edit: Or better yet, try and figure it out for yourself. The tools to do this analysis are available to everyone.
- promptdaddy 6 months agoWhat are these tools you speak of ?
- andrewflnr 6 months agoDude, I already have way too many projects. I'm going to wait for the experts on this one. :D
- promptdaddy 6 months ago
- ababaian 6 months ago
- kettleballroll 6 months agoHow did you find these things? Since you "stumbled upon them", you probably didn't know what you eee looking for, so... How did this research get started?
- pseudosudoer 6 months agoAny chance there is a correlation between Obelisks and autoimmune diseases?
- ababaian 6 months agoNot that I know of, although this doesn't mean it isn't happening, it's something which has to be investigated in detail.
- ababaian 6 months ago
- mmooss 6 months agoThe OP addresses it to some degree, but how does this fit with the fundamental classes of phylogeny?
My poor understanding has been that there are cellular organisms / 'biota' (if those are the right terms - prokaryotes, eukaryotes, etc.) and viruses. Where do obelisks, Virus-Viroid Hybrids, Zeta-Elements, Ambiviruses all fit in that scheme, if they do at all? Or is there a new scheme?
And it is very cool for you to answer questions here. Remember us if you visit Sweden someday! :)
- ababaian 6 months agoThe classic phylogenetic classes are a fantastic model, but https://en.wikipedia.org/wiki/All_models_are_wrong :)
Phylogeny is the study by which things relate to one another. There is a divergence point at which point it becomes impossible to relate two sequences to one another. Obelisks, Zeta-Elements, Deltaviruses, viroids all veer towards their own divergence point into infinity, but their are higher-order genome organization traits which are consistent. We don't know if these traits are the same by origin, or the same by chance. Interestingly Ambiviruses also have this genome organization, but they have a protein which is de facto of an RNA virus.
My opinion is that these simple genome layouts (structured circular RNA elements with ribozymes) are like a cauldron of mixing simple genes, and when they come together just the right way, we see those lineages take off. Think of it as an ocean of ancient primordial RNA replicators, ready to fire off, and this process is ongoing even today.
- mmooss 6 months agoThank you.
> The classic phylogenetic classes are a fantastic model, but https://en.wikipedia.org/wiki/All_models_are_wrong :)
For the record, yes, I know. Unfortunately, we need models to organize the world in our limited brains, and the less intimate experience one has of something, the more simplified their model. This isn't my day job! :)
- mmooss 6 months ago
- ababaian 6 months ago
- dj_gitmo 6 months agoHow did you come up with the name? And is the “Oblin” protein is named after Obelisks?
- ababaian 6 months agoVanya thought that when you run them through the RNA folding software, it would give you these unusual straight rods which reminded him of Cleopatra's Needle (Obelisk, https://en.wikipedia.org/wiki/Cleopatra%27s_Needles). The name stuck around. Incidentally the name has some cool 2001: Space Odyssey monolith vibes to it, which I think has been fitting.
- CRConrad 6 months ago> unusual straight rods which reminded him of Cleopatra's Needle (Obelisk ...
Also, of course, Obelix[1], purveyor of menhirs[2] -- Breton "standing stones".
- brian-armstrong 6 months agoDoesn't a name like this risk sensationalizing the discovery? I mean it's interesting to me as a layperson, but "obelisks" in pop culture carry a lot of woo factor
- CRConrad 6 months ago
- ababaian 6 months ago
- RobotToaster 6 months agoCan we kill them? Do we know if any drugs have anti-obelisk effects?
- ababaian 6 months agoWhat have Obelisks ever done to you? Our first reaction shouldn't be to kill everything we don't understand :'(
Good pragmatic question though. It's not clear if any drugs up- or down-regulate Obelisk genome copy, you could re-investigate other drug-treatment studies to see if Obelisks incidentally present are altered and get an "accidental" study.
From a molecular perspective, the most likely compounds and methods would be those which work against viroid replication (i.e. RNA polymerase inhibitors, translational inhibitors, CRISPR,...). You just have to maintain a preferential toxicity to Obelisks over host cells.
- hinkley 6 months agoWell, you say they aren’t linked to IBS or Crohn’s, but there’s a difference between proving that they don’t all cause intestinal issues and eliminating every strain as a cause. It’s kind of embarrassing how recently h pylori was caught red handed.
Some people carry staph on their skin their whole life and never end up with a lesion.
- hinkley 6 months ago
- ababaian 6 months ago
- zaptheimpaler 6 months agoThis looks really cool, im interested in learning more about computational biology, RNA and would like to work as a dev to get exposure. Is there an opportunity to gain research experience/skills as a developer in your lab? Could you provide a rough range of compensation as well?
- adrian_b 6 months agoThe free preprint:
- kaycebasques 6 months agoI think I've missed the AMA window but I'll try anyways.
Presumably you've got a lot of follow-up research to do. What are the most important research questions re: obelisks now?
- ghostly_s 6 months agoI’m not gleaning what observations support these things being alive from the article-how do we know they're not just an odd form of garbage RNA produced by cells?
- ababaian 6 months agoThat's exactly what they are though, some piece of garbage RNA cells are producing. A lot of things meet that definition if you think about it.
- ababaian 6 months ago
- Communitivity 6 months agoThis is fascinating work and sounds a little like the research my daughter says she wants to study (she's only a sophomore at UMBC right now, though). She hopes to get an internship in the summer of her junior year. She is interested in plant biology and bioengineering.
If I understand correctly, plants have RNA - would this mean new RNA-based lifeforms could also be found within plants?
- light_hue_1 6 months agoThere are no RNA-based lifeforms. All known life is DNA based but uses RNA internally. The earliest common ancestor of all life was DNA based, LUCA https://en.wikipedia.org/wiki/Last_universal_common_ancestor
There is a hypothesis that once upon a time life passed though an RNA only stage without DNA and proteins. RNA world https://en.wikipedia.org/wiki/RNA_world
- ababaian 6 months agoI like to view it that we're all RNA-based lifeforms. Operationally: DNA, RNA, or other are just a vehicles which hold our information.
This podcast RadioLab with Carl Zimmer (11m) I think captures the essence of the idea near perfectly: https://radiolab.org/podcast/creation-translation
- lolinder 6 months agoThe premise of TFA is that we're treating viruses and virus-like things as a class of lifeform. There are RNA viruses, and these obelisk things are also RNA-based. Presumably that's what OP is asking about with regard to RNA-based lifeforms.
- cyberax 6 months ago> There are no RNA-based lifeforms.
There are viruses that have entirely RNA-based lifecycle (even using RNA-dependent RNA polymerase). Our very favorite COVID virus is one of them.
- ababaian 6 months ago
- light_hue_1 6 months ago
- alexwasserman 6 months agoYou say human a lot, but are they really unique to humans, or we just haven’t looked at other animal definitive tracts yet?
- endofreach 6 months agoSerious question, super OT but i've wanted to ask this someone who works non-trivial fields like you for a long time: Have you ever used a LLM for something relevant to your work?
- ababaian 6 months agoIt's not OT at all. It depends on what you mean exactly by LLM, but we use them all the time. ESMFold2 was an LLM [https://www.science.org/doi/10.1126/science.ade2574] and was instrumental in advancing deep protein fold prediction in metagenomic space. Likewise AlphaFold2 it's direct application in creating FoldSeek for ultra-deep homology search. Both of these actually have radically improved our capacity to say Obelisks are _REALLY_ not like anything that's known.
Besides that I encourage all students to use ChatGPT for research, coding, copy editing, etc... I haven't encountered an LLM that can deal with difficult domain problems like we're facing, but I'd welcome the help. I'm for using all tools available, my main criticism with AI/LLM in general is the poor way in which uncertainty is reported.
- ababaian 6 months ago
- aaroninsf 6 months agoThis is by a good margin the most interesting knowledge-domain I have encountered in at minimum weeks. Super-interesting.
Hammerhead Self-Cleaving Ribozyme is quite a chunk of English.
- ucha 6 months agoDoes this support the Selfish Gene theory of Richard Dawkins? They look like the smallest self-replicating molecules that he mentions in the beginning of the book.
- ababaian 6 months agoAbsolutely, reading Selfish Gene in high school set me on the path to this type of exploration. Genes, in the pure abstract sense, are the unit by which we interrogate understand evolutionary change. There's a large grey area about the boundaries of genes, but after a certain point, genes assemble into operational units larger than themselves, a genome. Obelisks are some of the simplest, most rudimentary genomes described thus far.
- mensetmanusman 6 months agoIt’s amazing how genes choose from the chaos if 10^200 possibilities and don’t self destruct.
- mensetmanusman 6 months ago
- ababaian 6 months ago
- Sxubas 6 months agoThanks for your work on the research. I get a feeling of amusement and wandering when thinking on what functions the obelisks may have.
I am hopeful this discovery can lead to technology to improve people's life. Just thinking out loud, cancer treatments, orphan diseases treatment, prevent Alzheimer's progression, new vaccines.
Very long shots, but that's the beauty of unknowns. I'm highly jealous of scientists that will formulate and test hypothesis around this topic.
- ababaian 6 months agoEveryday I get to do this kind of research, I'm grateful to be the one doing it.
- ababaian 6 months ago
- marojejian 6 months ago
- andybak 6 months ago> researchers have stumbled across what seem to be an entire new class of virus-like objects.
I was confused at first. This isn't "Class" in the technical sense (i.e. the level between Phylum and Order)
- treprinum 6 months agoI am really glad we are finding new pieces of the puzzle of how our gut works and perhaps can someday understand their effect on immunity, neuro-degeneration, cancer etc. for which we now only have accidental findings.
- readyplayernull 6 months ago> Obelisks' genetic sequences are only around 1,000 characters
So they have a higher chance of being re-created by random chemical processes at mostly any point in time and place in the universe. Omniterrestrial?
- asymmetric 6 months agoThe study hasn’t been peer reviewed yet, so the title is a bit too confident in its claim.
- moralestapia 6 months agoHere's the journal publication (which came later during the year):
https://www.cell.com/cell/abstract/S0092-8674(24)01091-2
It truly is a new class of genomic elements.
- kelseyfrog 6 months agoAnd it's been replicated?
- moralestapia 6 months agoIt's using data from the Integrative Human Microbiome Project [1], so, in a sense, hundreds of high-quality biological replicas support their findings. Obelisks were found in a substantial portion of them.
They then expanded the search to millions of sequences, which are publicly available, and found ~30k different classes(!) of Obelisk elements. One could argue that the quality of each of these "experiments" may not be as good as IHMP, but still, the signal is more than sufficient to clearly demonstrate the existence and implied significance of these elements.
- moralestapia 6 months ago
- kelseyfrog 6 months ago
- robertlagrant 6 months agoMaybe Obelisks need an asterisk until peer review?
- readthenotes1 6 months agoIn case others don't get it:
- lifeisstillgood 6 months agoWe should not be so dogmatix in requiring scientific protocol - maybe there are some VitalStatistix we are not yet aware of.
- fecal_henge 6 months agoSplendido.
- readthenotes1 6 months ago
- tokai 6 months agoPeer review doesn't do what you think it does.
- martin82 6 months agoco-author (top comment?) just said that it is peer reviewed now.
Not that it matters at all.
- moralestapia 6 months ago
- dbcooper 6 months agoArticle is from January. Has the study now been published in a journal?
- magicalhippo 6 months agoYes, the bioRxiv entry[1] links to Cell[2].
[1]: https://www.biorxiv.org/content/10.1101/2024.01.20.576352v1
- Thoreandan 6 months agoFrom one of the paper's authors elsewhere in this thread: https://www.cell.com/cell/abstract/S0092-8674(24)01091-2
- magicalhippo 6 months ago
- kylehotchkiss 6 months agoBut how are these life? Life generally defined by ability to reproduce and these would be piggybacking on hosts cellular machinery.
Do these have a known utility or is it possible some junk DNA is involved with their encoding?
- joe_the_user 6 months agoMy crude understanding is the definition of life has gotten quite fuzzy in current biological theory.
Notably, I think the "viruses first" theory for the origin life has gained force. This says that first came protein/DNA soup, then came viruses and only then came cellular organisms.
And if you want something that doesn't "piggyback", you'd have to wait for photosynthesizing plants and that's several steps further in evolution (in my layman's understanding of current theory).
- joe_the_user 6 months ago
- dr_dshiv 6 months agoRemnant of RNA world?
- ababaian 6 months agoAren't we all?
- ababaian 6 months ago
- husamia 6 months agoit's interesting that RNA modifications are more diverse than DNA and we are just starting to develop ways to discover them. Nanopore sequencing technology from Oxford Nanopore Technology is the first technology that can sequence native RNA and their modifications. Have you explored this area?
- casenmgreen 6 months agoThis is a staggering and amazing finding.
- tglobs 6 months agoWhy call these obelisks their own class instead of categorizing them as another type of virus?
- casenmgreen 6 months agoPrivacy policy / cookie dialogue on that page is class A scum-worthy dark pattern.
- jeroen 6 months agoI'm more surprised when I encounter one that isn't hostile like this.
- jeroen 6 months ago