4 hours ago
This Google blog post is a distilled version of a Deep Mind blog post:
https://deepmind.google/blog/alphagenome-atlas-a-predictive-...
They are only announcing a cache. The origin for the cache is not discussed.
In particular, the question of whether to trust the predictions is not addressed. For that, I think the citation is from January:
an hour ago
People are upvoting this because it has the “Alpha______” prefix. Meanwhile, everyone in the field of genomics knows that AlphaGenome provides essentially zero improvements over the previous SOTA, Borzoi…
an hour ago
> a database that predicts the effects of every possible single nucleotide variant in the human genome. We used the AlphaGenome AI model to pre-calculate the regulatory impact of all 9 billion single-letter genetic changes, resulting in a massive, 1-petabyte dataset.
This is for a database, no? While Borzoi is a model?
> Here, we introduce Borzoi, a model that learns to predict cell-type-specific and tissue-specific RNA-seq coverage from DNA sequence.
20 minutes ago
> Meanwhile, everyone in the field of genomics knows that AlphaGenome provides essentially zero improvements over the previous SOTA, Borzoi…
Can you elaborate on this? I'm confused why Google would build something that provides zero improvements over SOTA, Borzoi... as you mention. I'm not familiar with this field, just curious.
4 hours ago
Videos. [2] is for the scientists to start using AlphaGenome Atlas from AntiGravity.
2 hours ago
Not a word about promoter sequences.
Imagine cellular activity as an industry zone, its not just what you can or can not make, its also 'for what concentrations of chemical species, what transcription rates should be used' so apart from the discrete Mendelian aspects (like what eye color or what have you) there is also a concensus sequence and deviations from consensus. They mention the dataset captures non-coding DNA, which should imply promoter sequences. Will it be possible to query the atlas for joint probabilities of promoter and putative target protein occurence in human genomes?
Personalized medicine could never credibly take off as long as promoter sequences were excised before sequencing!
3 hours ago
Can this be used with a 23andMe genome to find pathogenic mutations?
2 hours ago
23andMe and similar companies don't transcribe your entire genome because that would cost way more than they charge you. They just sample a few tiny sections of it.
3 minutes ago
fwiw sequencing your entire genome only costs $400 or so from providers like sequencing.com
25 minutes ago
A few as in tens of thousands.
23andMe used a custom Illumina Infinium microarray designed around segments of particular interest.
18 minutes ago
Probably not any 23andMe haven't already told you about. They test a limited set of SNPs, balancing between ones thought useful for genealogy, ones useful for ethnicity estimates and ones thought useful for health-related things (the latter they would like to make their main selling point, the two former are really all commercial DNA services' bread and butter).
It's unlikely that they would luck into testing some unknown SNP which turned out to be relevant for disease.
an hour ago
23andMe tests SNP's (single nucleotides) that are inferred to be significant in protein function/epigenitics.
Those SNP's i believe are testd from primers
so what 23andMe does is specifically on the back of previous research and afaik their data isnt technically clinically significant as most findings need confirmation or more tests.
3 hours ago
Not really, no.
2 hours ago
Why not? There is no logical reason as to why this would not work, IF it works in the first place, which I don't know. In theory the problem space here is finite, so there is of course a way to predict everything. Whether this is the case right now - who knows; I probably don't think it is currently ready. But eventually it will be. And it should not be in the hands of private companies.
2 hours ago
So it sounds like you're coming to this with very little knowledge about biology. I encourage reading up on modern challenges in pathogenic prediction, especially with regards to SNPs: on their own, with the exception of a few diseases, individual SNP predictions are meaningless in terms of actual pathogenicity.
2 hours ago
Because 23andme does not sequence your genome, only substring matches linked to specific gene variants.
4 hours ago
I really love Deep Mind, its genuinely focused on using AI to make the world a better place.
4 hours ago
Be careful when loving shareholder-driven endeavors; they're a single executive decision away from breaking your heart.
4 hours ago
It's "shareholder-linked" (probably not "driven"), but yes, upvoted. Nonetheless, even when with grave faults in the path, Google has managed to give us marvellous, paramount free resources (Google Street, Google Museum...), and would have given more (Google Books - if the negotiation had succeeded). I am grateful.
3 hours ago
Google Streetview/maps isn't even the best mapping/'streetview' tool at this point. Filled with ads and bugs, some major bugs have been there for 10+ years.
3 hours ago
Now I'm curious. What is the better mapping and streetview tool at this point?
41 minutes ago
For maps there's quite a few, just anything that doesn't have the amount of ads beats it. For Streetview apple lookaround is a better experience if it's mapped well in your area.
I believe Google has been trying to improve their Streetview in coverage as apple has become a concurrent, but it still has the same bugs people have been complaining about for years (like going backward when pressing forward type of bad). If I remember correctly the bugs have been 'closed' or something before just to include it in the same slightly different way.
2 hours ago
Organic Maps for maps
2 hours ago
Why would you be grateful? I don't think any of that should be under private control to begin with.
23 minutes ago
And which public enterprise has undertaken those projects?
Looking at current hot problems, and exemplary of the situation: which public entities have endeavoured to backup the most important channels and videos from YouTube, the current historic video archive?
Why would I be grateful?! Because somebody did create those resources and made them available! And they did to common good (if anybody refused the expression "for" common good).
3 hours ago
Be careful when loving government/philanthropy/ngo-driven endeavors; they're a single executive decision away from breaking your heart.
4 hours ago
No AI company is doing that. There is some benefits but that’s not what any AI for profit will ever focus upon
4 hours ago
DeepMind objectively is doing that. The world is not so black and white as "evil profit-hungry corporations will never do anything good".
4 hours ago
I preemptively acknowledged that in my comment…
3 hours ago
If Google does anything good for the world it is entirely accidental, and I am certain they will correct this mistake later.
an hour ago
I think they have done some quantifiable good.
Alphafold or Earthquake alerts for example spring to mind.
You could argue about comprehensive "free" services (search, Gmail, YouTube, docs, etc) and "free" browsers and "free" operating systems (yeah plural) being net positives in the world but I know that many people view these as negatives as they (rightly or wrongly) view them as vehicles to sell ads or collect data, but I bet you 100 bucks the typical person on the street using their android phone to search Google, watch YouTube, check their Gmail, write a doc, view photos etc doesn't give a shit about that and values these things - google has ended up as a monopoly as these things are actually pretty useful to people and thus valued, even if we in the tech world disagree with the adtech and data aspects. This monopoly was not a fait acompli or pure market abuse - Nokia, MS, flikr, Hotmail, yahoo etc were all incumbent and market dominant but are now irrelevant because the Google products were viewed as better by end users.
3 hours ago
Market reward is the best proxy for good we have.
What's your alternative? Executive fiat? Mob rule?
Deepmind is objectively good. These endeavors would always cost money. Better private capital than state, no one has to give Google anything. The state is not so generous.
2 hours ago
Mob rule, aka Democracy.
2 hours ago
Yes, we can just vote ourselves largess from the public purse.
2 hours ago
Not what democracy is.
Every single serious theorist of democracy has recognized this.
Democracy is about rule of law and institutions, not simply the mob.
Tyranny of the majority is the paramount concern of all democracies. Nevertheless, institutions can fail.
The idea that democracy always leads to desirable outcomes is evidently false by historical record.
I just don't understand people who think the state is better than Google. Google pretty much works. Google does not have police powers. Google cant surround your car and point guns at your family. Most of the bad things about Google are actually forced upon them by the state. Who could have even a passing knowledge of history and think states are somehow safer than corporations?
Take the United States. Should Donald Trump get a say in how to run Google? Maybe you like him, so how about the guy that comes after him, or the next one after that?
The American mob seems intent on leading their country into the gutter. Good thing (despite what the mob has thrown at it) that the United States still has strong institutions and strong markets. Without Silicon Valley, America would have already lost its place on the world stage.
But please lets hear more indictments from a population who can't even agree with itself on the issues already under the government's purview. Now let's set a precedent for even more state interference in the market, that way when your party (whatever side you are on) chokes the next future election away the next nut you don't agree with has even more power? Is that your plan?
21 minutes ago
> Should Donald Trump get a say in how to run Google?
Trump does have a say in how to run google, in a very real sense.
> The American mob seems intent on leading their country into the gutter. Good thing (despite what the mob has thrown at it) that the United States still has strong institutions and strong markets. Without Silicon Valley, America would have already lost its place on the world stage.
American elites are more than happy to follow basically whatever Trump wants them to do with the knowledge that they'll get preferential treatment by the law and win all the no-bid contracts he can award. But ask yourself if this would be any different if Trump were head of a large company under a smaller regulatory state. Seems pretty obvious that not much would change.
If by "the mob" you mean, a small cohort of wealthy, powerful, unscrupulous people, those people who make decisions that the rest of us have no say in, then I agree, that is a problem.
> But please lets hear more indictments from a population who can't even agree with itself on the issues already under the government's purview. Now let's set a precedent for even more state interference in the market, that way when your party (whatever side you are on) chokes the next future election away the next nut you don't agree with has even more power? Is that your plan?
In a smaller state, Trump would have continued to be an executive and the shape of his impact on the world might be slightly different, but would be no less oppressive. Ask tenants in NYC in the 1980s if they think a Donald Trump with smaller regulatory oversight who stayed in business would be better or worse for our country.
Do you think our country flips parties every year because our government does a good job of reflecting the will of the people?
2 hours ago
> Market reward is the best proxy for good we have.
Only if the goal is profit. All else be damned.
> What's your alternative? Executive fiat? Mob rule?
It is always bewildering to me that people see the world being raped by corporations and their negative externality, and they think that this is the pinnacle that humanity can achieve.
Nothing else can be better, if all your options are "mob rule" or "executive fiat".
I fancy myself a misanthrope, but damn, next to people in this forum I am probably the last humanist walking through the ruins of civilization.
> Deepmind is objectively good.
Google is objectively bad, and the world would be a much nicer place if they ceased to exist.
an hour ago
> It is always bewildering to me that people see the world being raped by corporations and their negative externality, and they think that this is the pinnacle that humanity can achieve.
> I fancy myself a misanthrope, but damn, next to people in this forum I am probably the last humanist walking through the ruins of civilization.
A humanist who uses a term of sexual violence so cavalierly?
If you are in the United States and you want a good civic lesson in state power, you should go to your nearest metropolitan criminal court. Maybe you will have a more accurate view of what government means then. At the very least, you will likely learn to banish such a casual invocation of that word from your lexicon.
> Nothing else can be better, if all your options are "mob rule" or "executive fiat".
What we have now is the middle ground between king and mob. It took centuries and much carnage to achieve. Does the famous Churchill quote really need to keep being repeated? Do you have an alternative proposal? Or maybe you have a sensible proposal that falls within the established democratic bounds? Or perhaps instead by "would be a much nicer place if they ceased to exist" we should simply infer a generalized whining rather than any specific policy prescription?
What's your actual idea, do you have one?
Maybe if you endeavored to communicate without self-satisfied hyperbole you would find we actually have reason to agree with each other.
2 hours ago
Executive fiat, given the right executive body, and undergirded by healthy autarkies or other cell-type self-direction, worked well in most locales for most of the years humans have been humans. Certain flashy civilizations, Sumer, Greece, have tended to draw the spotlight incommensurately with their population sizes. But even these functioned much more like what I describe above than anything modern.
2 hours ago
In another HN thread about this AlphaGenome Atlas, someone has posted a link to:
https://www.science.org/content/blog-post/mutate-em-all-and-...
which comments the results of this study:
https://www.biorxiv.org/content/10.64898/2026.07.25.740675v1
That study has done in reality what the AlphaGenome Atlas does in fiction, but instead for a human they have done it for one of the simplest viruses.
So they have fuzzed the virus by mutating one by one each position of its DNA.
And various dedicated AI models all made poor predictions of the results of that experiment, which casts doubts about the value of the AlphaGenome predictive map.
A virus is much simpler than a human, but even for that simple virus the effects of most of the mutations could not be predicted. A half of the mutations had harmful effects, and for a half of those it is unknown for now why they were harmful.
For a human the uncertainty about the effects of a mutation will be far greater than for one of the simplest viruses.
an hour ago
Yep. Sequence-to-function models are still very limited. AlphaGenome Atlas, despite the flashy branding, is unlikely to provide significant benefit to researchers.
2 hours ago
This has Demis written all over it. There is a great video of him with AlphaFold chatting with the team about releasing some results, and he asked something like “what if we just do them all?”
Very excited to see that happen here.
an hour ago
I don’t think Demis played a big role in this. It was mainly Ziga Avsec who developed Enformer (the first actually decent sequence-to-function model), and then AlphaGenome.
2 hours ago
This has nothing to do with AlphaFold at all- not sure if you were implying that. (the scientific contriution is welcome, but it's not particularly significant)
2 hours ago
Why are you excited about this?
an hour ago
There are AI labs headed by marketing CEOs that fake metrics, have an utter disregard for humanity and make up "AGI is imminent" propaganda for their IPO, and then there are AI labs headed by actual scientists that do actual science for humanity without constantly trying to put themselves into the spotlight.
an hour ago
Interestingly the labs led by marketing CEOs have much better models (astra is A LOT better than gemini).
3 hours ago
So I will know which DNAs to change to become a wolverine! yay!
3 hours ago
This may not be anything new but it makes using several Google/DeepMind resources a lot less painful.
I'm comfortable programming but others who also do mol bio may be less so or may not recognize when Claude is going off the rails.
3 hours ago
Anyone know if you can use an indel VCF file with this?
36 minutes ago
could this be used with a nebula genomic sequence to find pathogenic sequences?
3 hours ago
I saw a really interesting talk by Katie Pollard at ISMB this year about the limitations of variant prediction.
The gist was, can existing variation provide enough context to infer impact of variation? The answer seemed to be no.
Kind of like how frontier LLMs need to ingest larger and large amounts of text to advance. We are going to need to leverage comparative data from other species, and likely tremendous amounts of laboratory mutagenesis experiments to actually make headway on variant prediction. Nature, as it stands, just doesn't have enough human variation.
an hour ago
That's an interesting statement: "The gist was, can existing variation provide enough context to infer impact of variation? The answer seemed to be no."
Is this saying that if we were to sequence every human being on the planet, we'd still be unable to explain some phenotype differences caused by variation simply becase there aren't enough humans/enough variation? Interesting, as that's the first time I've heard that claim, and it would suggest that we spend our time working on mechanistic models of variant to phenotype.
2 hours ago
Yes, and genetic variant generally do not act in isolation. We currently focus on the small additive effects of variants because we can with small sample sizes—and 1 million humans is marginal using a GWAS cohort to dive into epistasis. But these interaction effects among variants are critical. Now almost completely deprecated.
2 hours ago
I'm convinced AI labs are absolutely hallucinating with their random names. It doesn't even mean anything anymore. Can we go back to numbers?
an hour ago
hmm gate keeping the database to elitist institutions and private businesses, I'm excited for the future!
2 hours ago
All controlled by an adCompany.
That is outright scary. Science is being slurped up here.
3 hours ago
unleash gemini 4 stop saving dario
5 hours ago
not my field so I can't judge how useful it is but assuming this data can be used for drug discovery and their ToS limiting to non-commercial use only. Does DeepMind plan on selling this data to pharmaceutical companies?
4 hours ago
Individual SNP predictions are not useful for drug discovery. Pharma might license this, but mainly out of fear of missing out.
4 hours ago
Probably. I know Google has partnerships with Pharma: https://www.merck.com/news/merck-and-google-cloud-partner-to...
4 hours ago
My understanding is that they already are, via Isomorphic Labs
4 hours ago
It's a bigger deal for diagnostics than for drug development.
4 hours ago
Hopefully. Tired, as a shareholder, to see the giveaways that deepmind is doing.
3 hours ago
Yeah, screw enabling life changing medical research, won't someone please think of the investors' financial returns??
2 hours ago
You can do both. Release it, but ask for money.
4 hours ago
Astroturfing, eh? This is verbatim the same comment as another user in this thread. Strange...
4 hours ago
PunchTornado was first, fr2029 was second
4 hours ago
i made no such comment. reported for harassment.
3 hours ago
so you just happen to be posting in a thread where some random user said you made a comment but you made no comments whatsoever
3 hours ago
You deleted the verbatim comment after it was pointed out. One was in thread and one was a stand alone comment. I was curious which was the throwaway so clicked into both profiles
3 hours ago
Cool. But is this actually advanced biology, or just making predictions about biology more accurate? Those aren't the same thing.
4 hours ago
the Google DeepMind PR team can't catch a break, shame it makes no sense to me - if someone's in the field maybe they could explain to the rest of us if this is a big deal, just a PR move, or nah?
4 hours ago
Just wait a few months, if it is a big deal that will likely be obvious by then
4 hours ago
i like this way of thinking ;)
2 hours ago
Does lots of good to my anxiety level :)
4 hours ago
What are you even talking about? What do you mean catch a break? They’ve made massive contributions over the last decade.
3 hours ago
I think orliesaurus meant the PR team is unable to rest because the rest of Deepmind is so prolific and keep coming out with new things for them to announce
2 hours ago
This is exactly what I meant!
3 hours ago
I don’t think so. The tone of the comment seems derogatory.
2 hours ago
No, it's exactly what @dronebetter said
4 hours ago
Google doing work to uncover the pandoras box of genetics? How long before they shove this under the rug...
4 hours ago
An atlas of the human genome from the company whose other atlas still routes me into a lake.
2 hours ago
I am freaking out. This is a huge moment.
I don't want to drag the discourse away from this achievement, but I hate how this is announced with blatant corporate advertising (our internal model, here are the benchmarks, gpt astra TM yours now for the low low price of £200pcm). I just didn't think Navier-Stokes falling would be sponsored by McDonald's.
Still. I am crying right now. Navier-Stokes is solved.
2 hours ago
This post is not about Navier-Stokes. That is in a different thread. Glad you're cry-happy though.
2 hours ago
Wait, is NS solved? I thought I was just a singularity thing?
2 hours ago
There is an announcement by OpenAI https://news.ycombinator.com/item?id=49613262