1d:I[70616,["/_next/static/chunks/0kcf5dt6yxdox.js","/_next/static/chunks/1su5n1uv1ti4s.js","/_next/static/chunks/0gvke1wv6udfg.js","/_next/static/chunks/1dfrle3jgmlzh.js","/_next/static/chunks/162lgvduyr-fa.js","/_next/static/chunks/2gcnnopam8e9w.js","/_next/static/chunks/2g60iab5a1pkg.js"],"Breadcrumb"] 1e:I[90762,["/_next/static/chunks/0kcf5dt6yxdox.js","/_next/static/chunks/1su5n1uv1ti4s.js","/_next/static/chunks/0gvke1wv6udfg.js","/_next/static/chunks/1dfrle3jgmlzh.js","/_next/static/chunks/162lgvduyr-fa.js","/_next/static/chunks/2gcnnopam8e9w.js","/_next/static/chunks/2g60iab5a1pkg.js"],"ArticleGrid"] 1f:T1857,

The bottleneck in reading the human genome has never been sequencing it — it has been working out which of the roughly 9 billion possible single-letter changes actually do anything. Google DeepMind's answer, released on September 8, is to stop guessing one variant at a time and precompute all of them: AlphaGenome Atlas is a one-petabyte catalogue of predicted molecular effects, free for academic use and more than 30 times the size of the AlphaFold Database.

Key takeaways

AlphaGenome Atlas contains predicted molecular effects for about 9 billion single-nucleotide variants, covering both coding and non-coding regions of the human genome.
A new AlphaGenome Variant Impact (AVI) score condenses AlphaGenome and AlphaMissense predictions into one number so researchers can rank variants before running experiments.
Early collaborators used it to pin down a DNM1 variant tied to epileptic encephalopathy and to surface 22% more non-coding trait associations in UK Biobank data.

Why this is a release at all

AlphaGenome itself shipped last year and is already in research use, a point DeepMind genomics lead Ziga Avsec acknowledged in a press briefing covered by The Verge, where he attributed the year-long gap to the raw cost of precomputing and analysing a variant space that large. The engineering work, in other words, was the product: an on-demand inference problem became a lookup table, which is a different thing to hand a biologist.

That distinction is what makes the AlphaFold comparison load-bearing. The 2022 database expansion took protein structures from roughly 190,000 experimentally determined entries to more than 200 million predictions, and the reason it reshaped the life sciences was accessibility rather than novelty — researchers with no coding background could finally browse it.

What a single variant lookup returns

Each of the 9 billion entries carries thousands of molecular effect predictions across hundreds of human and mouse cell types and tissues, which the AVI score compresses into one rankable number by folding in AlphaMissense's protein-impact predictions alongside AlphaGenome's regulatory ones. Feature attributions then decompose that score back into chromatin accessibility, splicing and conservation, so a high rank comes with a mechanism attached rather than just a warning light.

A fourth resource maps more than 2,500 recurrent DNA motifs — the short sequences that transcription factors bind. DeepMind reports best-in-class AVI performance on several variant pathogenicity and rare disease benchmarks, and stresses that the score works in non-coding DNA, the 98% of the genome that houses most trait-associated variants and where interpretation has always been weakest.

What collaborators found before launch

Working with the GREGoR Consortium, Broad Institute researchers Laura Covill and Anne O'Donnell-Luria used AVI to re-rank rare disease candidates that earlier analyses had passed over, and surfaced a DNM1 variant — a gene tied to epileptic encephalopathy — that the model predicted would create a faulty splice site and an abnormally extended protein. Experimental screens confirmed the prediction and turned up neighbouring variants behaving the same way.

A second test ran at population scale. Gareth Hawkes of the University of Exeter applied the Atlas to whole-genome data from over 54,000 UK Biobank participants and, by grouping rare variants according to predicted molecular effect, pulled 22% more non-coding associations out of the statistical noise — among them regulatory variants governing PLA2G7 and EGLN1. Narrowing a body-mass-index analysis to the most impactful 1% of non-coding variants left 19 genetic regions worth targeting experimentally.

Availability and limits

Access runs through the website portal, the AlphaGenome API and a skill inside Google Antigravity today for non-commercial use, with Google Cloud availability for commercial users promised later. DeepMind is blunt that this is a machine learning output and not clinical guidance — AlphaGenome has not been validated or approved for any clinical use, and every hit still needs wet-lab confirmation.

The release also fits a pattern. Google DeepMind's science group has been converting research models into operational infrastructure at a steady clip, most recently when WeatherNext bought forecasters an extra day of cyclone warning, and it arrives as co-founder Demis Hassabis steps back from running the lab to concentrate on scientific research and drug discovery at Isomorphic Labs.

FAQ

Is AlphaGenome Atlas free to use?

Yes, for non-commercial academic research. Access runs through a free web portal and the AlphaGenome API. Commercial use will be offered on Google Cloud at a later date.

What is the AVI score?

The AlphaGenome Variant Impact score combines AlphaGenome's regulatory predictions with AlphaMissense's protein-impact predictions into a single ranking number. It ships with feature attributions showing which molecular processes drive the score, and DeepMind reports best-in-class results across several variant pathogenicity and rare disease benchmarks.

Can doctors use it to diagnose patients?

No. DeepMind states that AlphaGenome has not been validated for clinical use and that Atlas output is not a substitute for professional medical advice, diagnosis, or treatment. It is a research prioritisation tool whose outputs still require experimental validation.

a:["$","div",null,{"className":"mx-auto max-w-7xl px-4 py-8 sm:px-6","children":[["$","script",null,{"type":"application/ld+json","dangerouslySetInnerHTML":{"__html":"{\"@context\":\"https://schema.org\",\"@type\":\"BreadcrumbList\",\"itemListElement\":[{\"@type\":\"ListItem\",\"position\":1,\"name\":\"Home\",\"item\":\"https://ainewsway.com\"},{\"@type\":\"ListItem\",\"position\":2,\"name\":\"#google-deepmind\",\"item\":\"https://ainewsway.com/tag/google-deepmind\"}]}"}}],["$","script",null,{"type":"application/ld+json","dangerouslySetInnerHTML":{"__html":"{\"@context\":\"https://schema.org\",\"@type\":\"ItemList\",\"@id\":\"https://ainewsway.com/tag/google-deepmind#itemlist\",\"name\":\"#google-deepmind Articles\",\"numberOfItems\":1,\"itemListElement\":[{\"@type\":\"ListItem\",\"position\":1,\"url\":\"https://ainewsway.com/articles/deepmind-precomputed-9-billion-dna-variants-into-a-1-petabyte-atlas-mu10uxvv\",\"name\":\"DeepMind Precomputed 9 Billion DNA Variants Into a 1-Petabyte Atlas\"}]}"}}],["$","$L1d",null,{"items":[{"label":"# google-deepmind"}]}],["$","h1",null,{"className":"mb-8 text-3xl font-bold text-zinc-900 dark:text-zinc-100","children":["#","google-deepmind"]}],[["$","$L1e",null,{"articles":[{"id":"9e8fcd93-26e3-4740-b10f-8771e4c9d88c","slug":"deepmind-precomputed-9-billion-dna-variants-into-a-1-petabyte-atlas-mu10uxvv","title":"DeepMind Precomputed 9 Billion DNA Variants Into a 1-Petabyte Atlas","subtitle":"AlphaGenome Atlas turns a per-variant inference problem into a free lookup table for academic researchers","content":"$1f","excerpt":"The bottleneck in reading the human genome has never been sequencing it — it has been working out which of the roughly 9 billion possible single-letter changes...","coverImage":"https://cdn.ainewsway.com/uploads/2026/09/b8c4bc45-dd47-43ab-be8a-fcb128442bb8.jpg","coverImageCaption":"An illustration of the DNA double helix, whose roughly 9 billion possible single-letter substitutions are now precomputed in Google DeepMind's AlphaGenome Atlas.","authorId":"6773f67d-d7fa-4b19-8f38-384f10ac761b","status":"PUBLISHED","publishedAt":"2026-09-14T09:13:36.416Z","viewCount":0,"likeCount":0,"readingTime":5,"isFeatured":false,"isBreakingNews":false,"metaTitle":"AlphaGenome Atlas Maps 9 Billion Human DNA Variants","metaDescription":"DeepMind's AlphaGenome Atlas precomputes molecular effects for 9 billion DNA variants in a 1-petabyte resource, free for academics. What it found so far.","metaKeywords":["AlphaGenome Atlas","Google DeepMind","AVI score","genomics AI","AlphaMissense","UK Biobank","rare disease","non-coding variants"],"ogImage":null,"aiSummary":"Google DeepMind released AlphaGenome Atlas on September 8, a one-petabyte catalogue of predicted molecular effects for roughly 9 billion single-letter DNA variants, free for academic use. It converts AlphaGenome from a per-variant inference tool into a browsable lookup resource, adding an AVI ranking score and a map of 2,500 DNA motifs. Early collaborators identified a DNM1 variant behind epileptic encephalopathy and surfaced 22% more non-coding associations in UK Biobank data. Commercial access on Google Cloud follows later.","aiTags":[],"sentiment":null,"articleType":"NEWS","videoEmbed":null,"videoType":null,"seoScore":85,"isAutoGenerated":true,"isBackfill":false,"pipelineStage":null,"copyrightScore":null,"similarityScore":null,"matchedSourceId":null,"createdAt":"2026-09-14T09:09:49.965Z","updatedAt":"2026-09-14T15:06:34.236Z","contentUpdatedAt":"2026-09-14T15:06:34.236Z","deletedAt":null,"author":{"id":"6773f67d-d7fa-4b19-8f38-384f10ac761b","name":"Seung Jung","username":"Seung Jung","avatar":null},"categories":[{"id":"e2b5fffc-de32-42aa-86fe-c75f219d6b4b","name":"AI & Machine Learning","slug":"ai-ml"}],"tags":[{"id":"07fdf649-dd9a-4a64-a675-a9507cc8af3f","name":"Google DeepMind","slug":"google-deepmind"},{"id":"09eb5399-3a10-4708-9092-4a4413b59ec3","name":"Google","slug":"google"},{"id":"145eb50b-4eb4-4cfa-8188-0f7463bc993a","name":"AI for Science","slug":"ai-for-science"},{"id":"8d186d2b-b0b2-490f-853f-6c0d8c6d89f1","name":"Genomics","slug":"genomics"},{"id":"a9083e67-1546-4fbe-9789-4e61748b6f9c","name":"AI Research","slug":"ai-research"}],"translations":[{"id":"48ce8950-14ea-48d0-9470-25e6658966c3","language":"ko","title":"딥마인드, DNA 변이 90억 개를 1페타바이트 지도에 미리 계산했다","subtitle":"AlphaGenome Atlas는 변이마다 추론하던 문제를 학술 연구자에게 무료로 열린 조회표로 바꿨다","excerpt":"인간 게놈을 읽는 데 병목이었던 것은 서열 해독이 아니라, 약 90억 가지 염기 한 글자 변화 가운데 실제로 무언가를 일으키는 것이 무엇인지 가려내는 일이었다.","metaTitle":"AlphaGenome Atlas, DNA 변이 90억 개 지도 공개","metaDescription":"DeepMind가 DNA 변이 90억 개의 분자 수준 영향을 미리 계산한 1페타바이트 자원 AlphaGenome Atlas를 학술용 무료로 공개했다. 초기 연구에서 무엇이 나왔나."},{"id":"6e8b885d-d603-4c47-bfdc-1d1578a6756b","language":"ja","title":"DeepMind、90億のDNA変異を1ペタバイトの地図に事前計算","subtitle":"AlphaGenome Atlasは変異ごとの推論問題を、学術研究者に無料で開かれたルックアップテーブルへ変えた","excerpt":"ヒトゲノムを読み解くボトルネックは配列決定ではなく、約90億通りある1文字の塩基置換のうち実際に影響を持つものを見極めることだった。","metaTitle":"AlphaGenome Atlas、90億のヒトDNA変異を地図化","metaDescription":"DeepMindのAlphaGenome Atlasは90億のDNA変異について分子レベルの影響を事前計算した1ペタバイト規模の資源で、学術利用は無料。これまでに何が見つかったか。"},{"id":"57476638-6b09-4008-a5da-35bcf3c9889d","language":"zh","title":"DeepMind预先算出90亿个DNA变异，汇成1PB图谱","subtitle":"AlphaGenome Atlas把逐个变异的推理问题，变成向学术研究者免费开放的查询表","excerpt":"解读人类基因组的瓶颈从来不是测序，而是判断约90亿种可能的单碱基改变中，究竟哪些真正起作用。","metaTitle":"AlphaGenome Atlas绘制90亿人类DNA变异图谱","metaDescription":"DeepMind的AlphaGenome Atlas预先计算了90亿个DNA变异的分子效应，形成1PB级资源并对学术用途免费开放。目前已有哪些发现。"},{"id":"dc81918a-2d4a-4c6c-9091-797d4a186f0f","language":"es","title":"DeepMind precomputó 9.000 millones de variantes de ADN en un atlas de un petabyte","subtitle":"AlphaGenome Atlas convierte un problema de inferencia variante a variante en una tabla de consulta gratuita para investigadores académicos","excerpt":"El cuello de botella para leer el genoma humano nunca fue secuenciarlo, sino averiguar cuáles de los cerca de 9.000 millones de cambios de una sola letra hacen algo.","metaTitle":"AlphaGenome Atlas mapea 9.000 millones de variantes","metaDescription":"AlphaGenome Atlas de DeepMind precomputa los efectos moleculares de 9.000 millones de variantes de ADN en un recurso de un petabyte gratis para académicos."}],"celebrities":[{"id":"bccb3daa-91a5-42d2-b58d-a641f2e3d534","nameKo":"구글 딥마인드","nameEn":"Google DeepMind","slug":"google-deepmind","profileImage":null},{"id":"42fd2fd4-e9ec-47e2-ba0c-4f8ccff5e2cc","nameKo":"데미스 하사비스","nameEn":"Demis Hassabis","slug":"demis-hassabis","profileImage":"https://ui-avatars.com/api/?name=Demis+Hassabis&size=400&background=84CC16&color=fff&bold=true"}]}]}],"$L20"]]}]

#google-deepmind

DeepMind precomputó 9.000 millones de variantes de ADN en un atlas de un petabyte

Suscríbete al Boletín

#google-deepmind

DeepMind precomputó 9.000 millones de variantes de ADN en un atlas de un petabyte