\n<\/p>\n
January 28, 2026<\/p>\n
3 min read<\/p>\n
Add Us On Google<\/span>Add SciAm<\/span><\/span><\/p>\n Google DeepMind unleashes new AI to investigate DNA\u2019s \u2018dark matter\u2019<\/p>\n DeepMind\u2019s AlphaGenome AI model could help solve the problem of predicting how variations in noncoding DNA shape gene expression<\/p>\n By Tanya Lewis edited by Claire Cameron<\/span><\/p>\n DNA is the blueprint for life, influencing everything about us\u2014including our health. We know that our genes, the genetic \u201cwords\u201d that encode proteins, play a major role in health and disease. But the vast majority of our genome\u2014more than 98 percent, in fact\u2014consists of DNA that doesn\u2019t build proteins. Once disregarded as \u201cjunk DNA,\u201d scientists now know that this molecular dark matter is crucial for determining gene activity in ways that keep us healthy\u2014or cause disease.<\/p>\n Exactly how this DNA shapes gene expression is a mystery\u2014but now the AI lab Google DeepMind has built a model that it says can predict the function of long stretches of noncoding DNA. The information it turns up could help solve the problem of predicting how these chunks of DNA influence our health.<\/p>\n Called AlphaGenome, the model takes in sequences of up to one million DNA letters, also known as base pairs, and predicts how mutations in those stretches affect gene expression. The model is described today in Nature. The tool, a version of which DeepMind has made freely available to other researchers, could help scientists narrow down theories for how certain DNA changes affect gene function. In turn, this knowledge could help scientists craft better treatments for genetic diseases.<\/p>\n If you’re enjoying this article, consider supporting our award-winning journalism by subscribing. By purchasing a subscription you are helping to ensure the future of impactful stories about the discoveries and ideas shaping our world today.<\/p>\n \u201cEver since the human genome was sequenced, people have been trying to understand the semantics of it\u2014this has been a longstanding goal for DeepMind,\u201d says Pushmeet Kohli, the company\u2019s vice president for science and a coauthor of the new study. \u201cIt\u2019s like you have a huge book of three billion characters and something wrong happened in this book.\u201d<\/p>\n \u201cAlphaGenome can be used to say, \u2018If you change these words, what would be the effect?\u2019\u201d he adds.<\/p>\n AlphaGenome works by combining information from several datasets focused on different aspects of gene expression\u2014how genes are turned on or off. The model is a successor of sorts to DeepMind\u2019s AlphaFold, an AI model that predicts the structure of almost every known protein from its amino acid sequence\u2014a central problem in biology. The researchers behind that effort shared the Nobel Prize in Chemistry in 2024. And in 2023 DeepMind released AlphaMissense, another AI tool that predicts how mutations in the regions of the genome that do generate proteins affect gene function.<\/p>\n AlphaGenome\u2019s developers say it performs as well or better than most other specialized models they tested. Previous tools generally required a trade-off between the length of a DNA sequence that could be used as input and accuracy. A key advance of AlphaGenome\u2019s approach is the ability to make accurate predictions about the function of extremely long genome sequences.<\/p>\n \u201cThe genome is like the recipe of life,\u201d Kohli said in a press briefing about the work. \u201cAnd really understanding \u2018What is the effect of changing any part of the recipe?\u2019 is what AlphaGenome sort of looks at.\u201d<\/p>\n AlphaGenome is a research tool\u2014it\u2019s not meant to be used clinically and its results can\u2019t be easily applied to individual humans. But it could have applications in understanding how the genome regulates genes in different types of cells or tissues. It could also help us understand diseases through massive genome-wide association studies or assist in studying cancer, because tumors can have many different genetic mutations, and it\u2019s not always clear which ones cause illness. The tool could even be useful for diagnosing rare conditions and designing new gene therapies.<\/p>\n \u201cFor all the best evaluations we have, AlphaGenome looks like they pushed [the field] forward a little bit,\u201d says David Kelley, a principal investigator at Calico Life Sciences, a subsidiary of Google\u2019s parent company Alphabet. Kelley was not involved with the study but has collaborated with the authors on a previous AI model. \u201cI think the long sequence length that they\u2019re able to work with here is definitely one of those major engineering breakthroughs,\u201d he says, adding that the new AI is \u201cincremental but real progress.\u201d<\/p>\n AlphaGenome has its limitations. It was trained on just two species\u2014humans and mice\u2014so isn\u2019t applicable to other species yet. And the tool might predict that a given DNA variant has no effect on gene expression when in fact it does.<\/p>\n Predicting how a disease manifests from the genome \u201cis an extremely hard problem, and this model is not able to magically predict that,\u201d says Žiga Avsec, a research scientist leading DeepMind\u2019s genomics initiative. But AlphaGenome can narrow down the pool of possible mutations involved in a disease, making it useful for prioritizing research to pinpoint which gene variants are actually causing problems, he says.<\/p>\n DeepMind\u2019s researchers acknowledge that the model is imperfect. The company\u2019s researchers are working to both boost what its predictive power is and better report how uncertain those predictions are.<\/p>\n If you enjoyed this article, I\u2019d like to ask for your support. Scientific American<\/span> has served as an advocate for science and industry for 180 years, and right now may be the most critical moment in that two-century history.<\/p>\n I\u2019ve been a Scientific American<\/span> subscriber since I was 12 years old, and it helped shape the way I look at the world. SciAm <\/span>always educates and delights me, and inspires a sense of awe for our vast, beautiful universe. I hope it does that for you, too.<\/p>\n If you subscribe to Scientific American<\/span>, you help ensure that our coverage is centered on meaningful research and discovery; that we have the resources to report on the decisions that threaten labs across the U.S.; and that we support both budding and working scientists at a time when the value of science itself too often goes unrecognized.<\/p>\n In return, you get essential news, captivating podcasts, brilliant infographics, can’t-miss newsletters, must-watch videos, challenging games, and the science world’s best writing and reporting. You can even gift someone a subscription.<\/p>\n There has never been a more important time for us to stand up and show why science matters. I hope you\u2019ll support us in that mission.<\/p>\n","protected":false},"excerpt":{"rendered":" January 28, 2026 3 min read Add Us On GoogleAdd SciAm Google DeepMind unleashes new AI to investigate DNA\u2019s \u2018dark matter\u2019 DeepMind\u2019s AlphaGenome AI model could help solve the problem of predicting how variations in noncoding DNA shape gene expression By Tanya Lewis edited by Claire Cameron DNA is the blueprint for life, influencing everything<\/p>\n","protected":false},"author":1,"featured_media":43078,"comment_status":"open","ping_status":"open","sticky":false,"template":"","format":"standard","meta":{"footnotes":""},"categories":[50],"tags":[22594,2430,19402,18793,2016,1765,5552,22593],"class_list":{"0":"post-43077","1":"post","2":"type-post","3":"status-publish","4":"format-standard","5":"has-post-thumbnail","7":"category-environment","8":"tag-alphagenome","9":"tag-dark","10":"tag-deepmind","11":"tag-dnas","12":"tag-google","13":"tag-investigate","14":"tag-matter","15":"tag-unleashes"},"_links":{"self":[{"href":"https:\/\/naijaglobalnews.org\/index.php?rest_route=\/wp\/v2\/posts\/43077","targetHints":{"allow":["GET"]}}],"collection":[{"href":"https:\/\/naijaglobalnews.org\/index.php?rest_route=\/wp\/v2\/posts"}],"about":[{"href":"https:\/\/naijaglobalnews.org\/index.php?rest_route=\/wp\/v2\/types\/post"}],"author":[{"embeddable":true,"href":"https:\/\/naijaglobalnews.org\/index.php?rest_route=\/wp\/v2\/users\/1"}],"replies":[{"embeddable":true,"href":"https:\/\/naijaglobalnews.org\/index.php?rest_route=%2Fwp%2Fv2%2Fcomments&post=43077"}],"version-history":[{"count":0,"href":"https:\/\/naijaglobalnews.org\/index.php?rest_route=\/wp\/v2\/posts\/43077\/revisions"}],"wp:featuredmedia":[{"embeddable":true,"href":"https:\/\/naijaglobalnews.org\/index.php?rest_route=\/wp\/v2\/media\/43078"}],"wp:attachment":[{"href":"https:\/\/naijaglobalnews.org\/index.php?rest_route=%2Fwp%2Fv2%2Fmedia&parent=43077"}],"wp:term":[{"taxonomy":"category","embeddable":true,"href":"https:\/\/naijaglobalnews.org\/index.php?rest_route=%2Fwp%2Fv2%2Fcategories&post=43077"},{"taxonomy":"post_tag","embeddable":true,"href":"https:\/\/naijaglobalnews.org\/index.php?rest_route=%2Fwp%2Fv2%2Ftags&post=43077"}],"curies":[{"name":"wp","href":"https:\/\/api.w.org\/{rel}","templated":true}]}}On supporting science journalism<\/h2>\n
It\u2019s Time to Stand Up for Science<\/h2>\n