{"id":41148,"date":"2021-07-22T17:00:20","date_gmt":"2021-07-22T15:00:20","guid":{"rendered":"https:\/\/www.embl.org\/news\/?p=41148"},"modified":"2024-03-22T11:22:50","modified_gmt":"2024-03-22T10:22:50","slug":"alphafold-database-launch","status":"publish","type":"post","link":"https:\/\/www.embl.org\/news\/science\/alphafold-database-launch\/","title":{"rendered":"DeepMind and EMBL release the most complete database of predicted 3D structures of human proteins"},"content":{"rendered":"\n<p>DeepMind today announced its partnership with the European Molecular Biology Laboratory (EMBL), Europe\u2019s flagship laboratory for the life sciences, to make the most complete and accurate database yet of predicted protein structure models for the human proteome. This will cover all ~20,000 proteins expressed by the human genome, and the data will be freely and openly available to the scientific community. The database and artificial intelligence system provide structural biologists with powerful new tools for examining a protein\u2019s three-dimensional structure, and offer a treasure trove of data that could unlock future advances and herald a new era for AI-enabled biology.<\/p>\n\n\n\n<p>AlphaFold\u2019s <a href=\"https:\/\/deepmind.com\/blog\/article\/alphafold-a-solution-to-a-50-year-old-grand-challenge-in-biology\">recognition<\/a> in December 2020 by the organisers of the Critical Assessment of protein Structure Prediction (CASP) benchmark as a solution to the 50-year-old grand challenge of <a href=\"https:\/\/en.wikipedia.org\/wiki\/Protein_structure_prediction#:~:text=Protein%20structure%20prediction%20(also%20called,structure%20from%20its%20primary%20structure.\">protein structure prediction<\/a> was a stunning breakthrough for the field. The AlphaFold Protein Structure Database builds on this innovation and the discoveries of generations of scientists, from the early pioneers of protein imaging and crystallography, to the thousands of prediction specialists and structural biologists who\u2019ve spent years experimenting with proteins since. The database dramatically expands the accumulated knowledge of protein structures, more than doubling the number of high-accuracy human protein structures available to researchers. Advancing the understanding of these building blocks of life, which underpin every biological process in every living thing, will help enable researchers across a huge variety of fields to accelerate their work.<\/p>\n\n\n\n<p>Last week, the methodology behind the latest highly innovative version of AlphaFold, the sophisticated AI system announced last December that powers these structure predictions, and its open source code were published in <em>Nature<\/em>. Today\u2019s announcement coincides with a second <em><a href=\"https:\/\/www.nature.com\/articles\/s41586-021-03828-1\">Nature<\/a><\/em> paper that provides the fullest picture of proteins that make up the human proteome, and the release of 20 additional organisms that are important for biological&nbsp; research.<\/p>\n\n\n\n<p>\u201cOur goal at DeepMind has always been to build AI and then use it as a tool to help accelerate the pace of scientific discovery itself, thereby advancing our understanding of the world around us,\u201d said DeepMind Founder and CEO Demis Hassabis, PhD. \u201cWe used AlphaFold to generate the most complete and accurate picture of the human proteome. We believe this represents the most significant contribution AI has made to advancing scientific knowledge to date, and is a great illustration of the sorts of benefits AI can bring to society.\u201d<\/p>\n\n\n\n<h2 class=\"wp-block-heading\"><strong>AlphaFold is already helping scientists to accelerate discovery&nbsp;<\/strong><\/h2>\n\n\n\n<p>The ability to predict a protein\u2019s shape computationally from its amino acid sequence \u2013 rather than determining it experimentally through years of painstaking, laborious and often costly techniques \u2013 is already helping scientists to achieve in months what previously took years.<\/p>\n\n\n\n<p>\u201cThe AlphaFold database is a perfect example of the virtuous circle of open science,\u201d said EMBL Director General Edith Heard. \u201cAlphaFold was trained using data from public resources built by the scientific community so it makes sense for its predictions to be public. Sharing AlphaFold predictions openly and freely will empower researchers everywhere to gain new insights and drive discovery. I believe that AlphaFold is truly a revolution for the life sciences, just as genomics was several decades ago and I am very proud that EMBL has been able to help DeepMind in enabling open access to this remarkable resource.\u201d<\/p>\n\n\n\n<p>AlphaFold is already being used by partners such as the <a href=\"https:\/\/dndi.org\/\">Drugs for Neglected Diseases Initiative<\/a> (DNDi), which has <a href=\"https:\/\/www.wired.co.uk\/article\/deepmind-alphafold-protein-diseases\">advanced their research into life-saving cures<\/a> for diseases that disproportionately affect the poorer parts of the world, and the <a href=\"https:\/\/www.port.ac.uk\/research\/research-centres-and-groups\/centre-for-enzyme-innovation\">Centre for Enzyme Innovation<\/a> (CEI) is using AlphaFold to help engineer faster enzymes for recycling some of our most polluting single-use plastics. For those scientists who rely on experimental protein structure determination, AlphaFold&#8217;s predictions have helped accelerate their research. For example, a team at the University of Colorado Boulder is finding promise in using AlphaFold predictions to study antibiotic resistance, while a group at the University of California San Francisco has used them to <a href=\"https:\/\/www.biorxiv.org\/content\/10.1101\/2021.05.10.443524v1\">increase their understanding of SARS-CoV-2 biology<\/a>.<\/p>\n\n\n\n<h2 class=\"wp-block-heading\"><strong>The AlphaFold Protein Structure Database<\/strong><\/h2>\n\n\n\n<p>The <a href=\"http:\/\/www.alphafold.ebi.ac.uk\">AlphaFold Protein Structure Database<\/a> builds on many contributions from the international scientific community, as well as AlphaFold\u2019s sophisticated algorithmic innovations and EMBL-EBI\u2019s decades of experience in sharing the world\u2019s biological data. DeepMind and EMBL\u2019s European Bioinformatics Institute (EMBL-EBI) are providing access to AlphaFold\u2019s predictions so that others can use the system as a tool to enable and accelerate research and open up completely new avenues of scientific discovery.<\/p>\n\n\n\n<p>\u201cThis will be one of the most important datasets since the mapping of the Human Genome,\u201d said EMBL Deputy Director General, and EMBL-EBI Director Ewan Birney. \u201cMaking AlphaFold predictions accessible to the international scientific community opens up so many new research avenues, from neglected diseases to new enzymes for biotechnology and everything in between. This is a great new scientific tool, which complements existing technologies, and will allow us to push the boundaries of our understanding of the world.\u201d<\/p>\n\n\n\n<p>In addition to the human proteome, the database launches with ~350,000 structures including 20 biologically-significant organisms such as E.coli, fruit fly, mouse, zebrafish, malaria parasite and tuberculosis bacteria. Research into these organisms has been the subject of countless research papers and numerous major breakthroughs. These structures will enable researchers across a huge variety of fields \u2013 from neuroscience to medicine \u2013 to accelerate their work.<\/p>\n\n\n\n<h2 class=\"wp-block-heading\"><strong>The future of AlphaFold<\/strong><\/h2>\n\n\n\n<p>The database and system will be periodically updated as we continue to invest in future improvements to AlphaFold, and over the coming months we plan to vastly expand the coverage to almost every sequenced protein known to science \u2013 over 100 million structures covering most of the UniProt reference database.<\/p>\n\n\n\n<p>To learn more, please see the <em>Nature<\/em> papers describing our <a href=\"https:\/\/www.nature.com\/articles\/s41586-021-03819-2\">full method<\/a> and the <a href=\"https:\/\/www.nature.com\/articles\/s41586-021-03828-1\">human proteome<\/a>, and read the <a href=\"https:\/\/deepmind.com\/research\/publications\/enabling-high-accuracy-protein-structure-prediction-at-the-proteome-scale\">Authors\u2019 Notes<\/a>. See the <a href=\"https:\/\/github.com\/deepmind\/alphafold\">open-source code to AlphaFold<\/a> if you want to view the workings of the system, and <a href=\"https:\/\/colab.research.google.com\/github\/deepmind\/alphafold\/blob\/main\/notebooks\/AlphaFold.ipynb\">Colab notebook<\/a> to run individual sequences. To explore the structures, visit EMBL-EBI\u2019s <a href=\"https:\/\/alphafold.ebi.ac.uk\/\">searchable database<\/a> that is open and free to all.<\/p>\n\n\n\n<h2 class=\"wp-block-heading\"><strong>Feedback from the scientific community<\/strong><\/h2>\n\n\n\n<p><strong>Jacques Dubochet, Nobel Laureate for Chemistry 2017, former Group Leader at EMBL<\/strong><br \/>&#8220;I love to know that the collaboration between&nbsp;DeepMind and EMBL will make all the knowledge&nbsp;about&nbsp;protein structure&nbsp;open to all.&#8221;<\/p>\n\n\n\n<p><strong>Paul Nurse, Nobel Laureate for Physiology or Medicine 2001, Director of the Francis Crick Institute and Chair of EMBL Science Advisory Committee<\/strong><br \/>\u201cWith this resource freely and openly available, the scientific community will be able to draw on collective knowledge to accelerate discovery, ushering in a new era for AI-enabled biology.\u201d<br \/><br \/><strong>Venki Ramakrishnan, Nobel Laureate for Chemistry 2009 and former President of the Royal Society<\/strong><br \/>\u201cThis computational work represents a stunning advance on the protein-folding problem, a 50-year-old grand challenge in biology.\u201d<br \/><br \/><strong>Prof. Dame Janet Thornton, Director Emeritus of EMBL-EBI<\/strong><br \/>\u201cThis contributes to our knowledge and understanding of living systems, with all the opportunities for humanity this will unlock.\u201d<\/p>\n\n\n\n<p>This post was originally published on <a href=\"https:\/\/www.ebi.ac.uk\/about\/news\/press-releases\/alphafold-database-launch\" data-href=\"insert original\" rel=\"canonical nofollow noopener noreferrer\" target=\"_blank\">EMBL-EBI News<\/a><\/p>\n\n\n<hr class=\"vf-divider\"\/>\n\n\n<h1 class=\"wp-block-heading\" id=\"a1\"><strong>DeepMind und EMBL ver\u00f6ffentlichen vollst\u00e4ndige Datenbank mit vorausberechneten 3D Darstellungen aller menschlichen Proteine<\/strong><\/h1>\n\n\n\n\n\n\n\n<p>DeepMind gab heute seine Partnerschaft mit dem Europ\u00e4ischen Laboratorium f\u00fcr Molekularbiologie (EMBL), Europas Vorzeigelabor f\u00fcr Biowissenschaften, bekannt, um die bisher vollst\u00e4ndigste und genaueste Datenbank der rund 20 000 Proteine und ihrer vorausberechneten Strukturen aus dem menschlichen Genom, also das vollst\u00e4ndige menschlichen Proteom, der wissenschaftliche Gemeinschaft frei zur Verf\u00fcgung zu stellen. Die Datenbank und die Anwendung k\u00fcnstlicher\/artifizieller Intelligenz (AI) bieten Strukturbiologen leistungsstarke neue Werkzeuge f\u00fcr die Untersuchung dreidimensionaler Strukturen von Proteinen. Sie er\u00f6ffnen damit eine Fundgrube an Daten, die k\u00fcnftige Forschung und Erkenntnisse erm\u00f6glichen und eine neue \u00c4ra der AI-gest\u00fctzten Biologie einl\u00e4uten k\u00f6nnen.<\/p>\n\n\n\n<p>Die <a href=\"https:\/\/deepmind.com\/blog\/article\/alphafold-a-solution-to-a-50-year-old-grand-challenge-in-biology\">Anerkennung von AlphaFold<\/a> als L\u00f6sung f\u00fcr die 50 Jahre alte gro\u00dfe Herausforderung der<a href=\"https:\/\/en.wikipedia.org\/wiki\/Protein_structure_prediction#:~:text=Protein%20structure%20prediction%20(also%20called,structure%20from%20its%20primary%20structure.\"> Proteinstrukturvorhersage<\/a> im Dezember 2020 durch die Organisatoren des CASP-Benchmarks (Critical Assessment of Protein Structure Prediction) stellte einen signifikanter Durchbruch in der Forschung dar. Die AlphaFold-Proteinstrukturdatenbank baut auf dieser Innovation und den Entdeckungen von Generationen von WissenschaftlerInnen auf: von den fr\u00fchen Pionieren der Proteinforschung und kristallographie bis hin zu den tausenden VorhersagespezialistInnen und StrukturbiologInnen, die seit Jahren Proteinstrukturen erforschen. Durch die Datenbank wird die Anzahl der hochpr\u00e4zise dargestellten menschlichen Proteinstrukturen, die Forschenden zur Verf\u00fcgung stehen, mehr als verdoppelt. AlphaFold erweitert damit das gesammelte Wissen \u00fcber Proteinstrukturen drastisch. Ein besseres Verst\u00e4ndnis dieser Bausteine des Lebens, die jedem biologischen Prozess in jedem Lebewesen zugrunde liegen, wird es Forschenden in den verschiedensten Bereichen erm\u00f6glichen, ihre Arbeit bedeutend zu beschleunigen.<\/p>\n\n\n\n<p>Letzte Woche wurde die Methodik hinter der neuesten, hochinnovativen Version von AlphaFold, dem hochentwickelten AI-System, das Strukturvorhersagen erm\u00f6glicht, und dessen Open-Source-Code in der wissenschaftliche Fachzeitschrift <em>Nature <\/em>ver\u00f6ffentlicht. Die heutige Ank\u00fcndigung f\u00e4llt mit einer zweiten Ver\u00f6ffentlichung in der <em><a href=\"https:\/\/www.nature.com\/articles\/s41586-021-03828-1\">Nature<\/a><\/em> zusammen, die nicht nur das umfassendste Bild der Proteine im menschlichen Proteom beschreibt, sondern auch die Proteinstrukturen 20 weiterer Organismen, die f\u00fcr die biologische Forschung bedeutend sind.<\/p>\n\n\n\n<p>&#8220;Unser Ziel bei DeepMind war es immer, AI zu entwickeln und sie dann als Werkzeug zu nutzen, um das Tempo wissenschaftlicher Entdeckungen zu beschleunigen und damit unser Verst\u00e4ndnis der Welt um uns herum zu verbessern&#8221;, sagte DeepMind-Gr\u00fcnder und CEO Demis Hassabis, PhD. &#8220;Wir haben AlphaFold verwendet, um das vollst\u00e4ndigste und genaueste Bild des menschlichen Proteoms zu erstellen. Wir glauben, dass dies der bedeutendste Beitrag ist, den AI bisher zum Fortschritt wissenschaftlicher Erkenntnisse geleistet hat, und es ist ein gro\u00dfartiges Beispiel f\u00fcr die Art von Nutzen, den AI der Gesellschaft bringen kann.&#8221;<\/p>\n\n\n\n<h2 class=\"wp-block-heading\"><strong>AlphaFold hilft WissenschaftlerInnen schon jetzt dabei, Entdeckungen zu beschleunigen&nbsp;<\/strong><\/h2>\n\n\n\n<p>Die M\u00f6glichkeit, die Form eines Proteins nun rechnerisch auf Grund seiner Aminos\u00e4uresequenz vorherzusagen &#8211; anstatt sie experimentell durch m\u00fchsame und oft kostspielige Techniken zu bestimmen &#8211; hilft WissenschaftlerInnen bereits jetzt, in wenigen Monaten zu erreichen, was fr\u00fcher Jahre dauerte.<\/p>\n\n\n\n<p>&#8220;Die AlphaFold-Datenbank ist ein perfektes Beispiel f\u00fcr den kontinuierlichen Erfolgszyklus von offener Wissenschaft&#8221;, sagte EMBL-Generaldirektorin Edith Heard. &#8220;AlphaFold wurde mit Datens\u00e4tzen aus \u00f6ffentlichen Ressourcen gef\u00fcttert, die von der wissenschaftlichen Gemeinschaft bereitgestellt wurden. Es ist also nur konsequent, dass seine Vorhersagen wiederum frei zug\u00e4nglich gemacht werden. Die offene und freie Weitergabe der AlphaFold-Vorhersagen wird es Forschenden \u00fcberall auf der Welt erm\u00f6glichen, neue Erkenntnisse zu gewinnen und Entdeckungen voranzutreiben. Ich gehe davon aus, dass AlphaFold eine wirkliche Revolution f\u00fcr die Biowissenschaften ist, so wie es die Genomik vor einigen Jahrzehnten war, und ich bin sehr stolz darauf, dass EMBL DeepMind dabei helfen konnte, den offenen Zugang zu dieser bemerkenswerten Ressource zu erm\u00f6glichen.&#8221;<\/p>\n\n\n\n<p>AlphaFold wird bereits von Partnern wie der <a href=\"https:\/\/dndi.org\/\">Drugs for Neglected Diseases Initiative (DNDi)<\/a> eingesetzt, die damit ihre <a href=\"https:\/\/www.wired.co.uk\/article\/deepmind-alphafold-protein-diseases\">Forschung nach lebensrettenden Heilmitteln f\u00fcr Krankheiten<\/a> vorantreibt, welche unverh\u00e4ltnism\u00e4\u00dfig stark wirtschaftlich benachteiligte Teilen der Welt betreffen. Das <a href=\"https:\/\/www.port.ac.uk\/research\/research-centres-and-groups\/centre-for-enzyme-innovation\">Centre for Enzyme Innovation<\/a> (CEI) wiederum nutzt AlphaFold, um schnellere Enzyme f\u00fcr das Recycling einiger der umweltsch\u00e4dlichsten Einwegkunststoffe zu entwickeln. Auch WissenschaftlerInnen, die ihre Forschung auf die experimentelle Bestimmung von Proteinstrukturen basieren, helfen die Vorhersagen von AlphaFold bereits dabei, ihre Arbeit zu beschleunigen. So nutzt zum Beispiel ein Forschungsteam an der University of Colorado Boulder die AlphaFold-Vorhersagen, um Antibiotikaresistenzen zu untersuchen, w\u00e4hrend eine Gruppe an der University of California San Francisco sie verwendet, um ihr Verst\u00e4ndnis der <a href=\"https:\/\/www.biorxiv.org\/content\/10.1101\/2021.05.10.443524v1\">Biologie von SARS-CoV-2<\/a> zu verbessern.<\/p>\n\n\n\n<h2 class=\"wp-block-heading\"><strong>Die AlphaFold-Proteinstruktur-Datenbank<\/strong><\/h2>\n\n\n\n<p>Die <a href=\"http:\/\/www.alphafold.ebi.ac.uk\/\">AlphaFold Protein Structure Database<\/a> baut sowohl auf vielen fr\u00fcheren Beitr\u00e4gen der internationalen wissenschaftlichen Gemeinschaft auf, als auch auf den hochentwickelten algorithmischen Innovationen von AlphaFold und der jahrzehntelangen Erfahrung des European Bioinformatics Institute (EMBL-EBI) &nbsp;bei der freien Bereitstellung von weltweit gesammelten biologischen Daten. DeepMind und EMBL-EBI stellen den Zugang zu den Vorhersagen von AlphaFold offen zur Verf\u00fcgung, damit die wissenschaftliche Gemeinschaft das System als Werkzeug dazu nutzen kann, weitere Forschung zu erm\u00f6glichen, zu beschleunigen und v\u00f6llig neue Wege der wissenschaftlichen Entdeckung zu er\u00f6ffnen.<\/p>\n\n\n\n<p>&#8220;AlphaFold wird sich als einer der wichtigsten Datens\u00e4tze seit der Kartierung des menschlichen Genoms herausstellen&#8221;, sagte der stellvertretende EMBL-Generaldirektor und EMBL-EBI-Direktor Ewan Birney. &#8220;Die AlphaFold-Vorhersagen der internationalen wissenschaftlichen Gemeinschaft offen zug\u00e4nglich zu machen, er\u00f6ffnet so viele neue Forschungsm\u00f6glichkeiten. Von vernachl\u00e4ssigten Krankheiten bis hin zu neuen Enzymen f\u00fcr die Biotechnologie und allem, was dazwischen liegt. Dies ist ein gro\u00dfartiges neues wissenschaftliches Werkzeug, das bestehende Technologien erg\u00e4nzt und es uns erm\u00f6glichen wird, die Grenzen unseres Verst\u00e4ndnisses der Welt zu erweitern.&#8221;<\/p>\n\n\n\n<p>Zus\u00e4tzlich zum menschlichen Proteom geht die Datenbank mit rund 350.000 weiteren Proteinstrukturen online, darunter Proteine von 20 biologisch bedeutsame Organismen wie der Fruchtfliege, der Maus, des Zebrafisches, des Malariaparasiten sowie von E.coli- und Tuberkulosebakterien. Die Forschung an diesen Organismen war bereits Grundlage zahlreicher bedeutender wissenschaftlicher Durchbr\u00fcche und entsprechender Ver\u00f6ffentlichungen. Ihre nun ver\u00f6ffentlichte Proteinstrukturen werden es Forschern in den verschiedensten Bereichen &#8211; von den Neurowissenschaften bis zur Medizin &#8211; erm\u00f6glichen, ihre Forschung schneller voranzutreiben.<\/p>\n\n\n\n<h2 class=\"wp-block-heading\"><strong>Die Zukunft von AlphaFold<\/strong><\/h2>\n\n\n\n<p>Die beteiligten Partner werden in den kommenden Monaten kontinuierlich in die Verbesserung von AlphaFold investieren und die Datenbank und das dahinterliegende System regelm\u00e4\u00dfig aktualisieren. So soll die Abdeckung auf fast alle sequenzierten Proteine, die der Wissenschaft bekannt sind, erweitert werden &#8211; \u00fcber 100 Millionen Strukturen, die den Gro\u00dfteil der <a href=\"https:\/\/www.uniprot.org\/\">UniProt-Referenzdatenbank<\/a> abdecken.<\/p>\n\n\n\n<p>Weitere Informationen k\u00f6nnen Sie dem Fachartikel zum Thema in der <em>Nature <\/em>entnehmen, der die <a href=\"https:\/\/www.nature.com\/articles\/s41586-021-03819-2\">vollst\u00e4ndige Methodik<\/a> und das <a href=\"https:\/\/www.nature.com\/articles\/s41586-021-03828-1\">menschliche Proteom<\/a> beschreibt sowie den <a href=\"https:\/\/deepmind.com\/research\/publications\/enabling-high-accuracy-protein-structure-prediction-at-the-proteome-scale\">Anmerkungen der Autoren<\/a>. Besuchen Sie den <a href=\"https:\/\/github.com\/deepmind\/alphafold\">Open-Source-Code zu AlphaFold<\/a>, wenn Sie sich die Funktionsweise des Systems ansehen m\u00f6chten und <a href=\"https:\/\/colab.research.google.com\/github\/deepmind\/alphafold\/blob\/main\/notebooks\/AlphaFold.ipynb\">Colab notebook<\/a>, um einzelne Sequenzen zu untersuchen. Um die Strukturen zu erforschen, besuchen Sie die <a href=\"https:\/\/alphafold.ebi.ac.uk\/\">Datenbank<\/a> von EMBL-EBI, die offen und kostenfrei zug\u00e4nglich ist.<\/p>\n\n\n<hr class=\"vf-divider\"\/>\n\n\n<h1 class=\"wp-block-heading\" id=\"a2\"><strong>DeepMind et l&#8217;EMBL publient la base de donn\u00e9es la plus compl\u00e8te de pr\u00e9dictions de structures 3D de prot\u00e9ines humaines<\/strong><\/h1>\n\n\n\n\n\n\n\n<p>DeepMind a annonc\u00e9 aujourd&#8217;hui son partenariat avec le Laboratoire europ\u00e9en de biologie mol\u00e9culaire (EMBL), le laboratoire de r\u00e9f\u00e9rence pour les sciences de la vie en Europe, afin de mettre \u00e0 disposition de mani\u00e8re libre et gratuite \u00e0 l\u2019ensemble de la communaut\u00e9 scientifique la base de donn\u00e9es la plus compl\u00e8te et la plus pr\u00e9cise \u00e0 ce jour de pr\u00e9dictions de mod\u00e8les de structures des prot\u00e9ines du prot\u00e9ome humain. Ceci \u00e9quivaut \u00e0 un ensemble d\u2019environ 20 000 prot\u00e9ines exprim\u00e9es par le g\u00e9nome humain. Ces donn\u00e9es seront mises \u00e0 disposition de la communaut\u00e9 scientifique de mani\u00e8re libre et gratuite. La base de donn\u00e9es et le syst\u00e8me d&#8217;intelligence artificielle fournissent aux chercheurs en biologie structurale de nouveaux et puissants outils pour examiner la structure tridimensionnelle d&#8217;une prot\u00e9ine, et offrent une v\u00e9ritable mine d\u2019informations qui pourrait d\u00e9bloquer de futures avanc\u00e9es et annoncer une nouvelle \u00e8re pour la biologie assist\u00e9e par l&#8217;intelligence artificielle.<\/p>\n\n\n\n<p>La <a href=\"https:\/\/deepmind.com\/blog\/article\/alphafold-a-solution-to-a-50-year-old-grand-challenge-in-biology\">reconnaissance<\/a> d&#8217;AlphaFold en d\u00e9cembre 2020 par les organisateurs de l&#8217;Evaluation critique de la pr\u00e9diction de la structure des prot\u00e9ines (ou CASP, <em>Critical Assessment of protein Structure Prediction, <\/em>en anglais) comme une solution au grand d\u00e9fi, vieux de 50 ans, de la <a href=\"https:\/\/fr.wikipedia.org\/wiki\/Pr%C3%A9diction_de_la_structure_des_prot%C3%A9ines\">pr\u00e9diction de la structure des prot\u00e9ines<\/a>, a constitu\u00e9 une remarquable avanc\u00e9e dans ce domaine. La base de donn\u00e9es sur la structure des prot\u00e9ines AlphaFold s&#8217;appuie sur cette innovation et sur les d\u00e9couvertes de g\u00e9n\u00e9rations de scientifiques, depuis les premiers pionniers de l\u2019imagerie et de la cristallographie des prot\u00e9ines, jusqu&#8217;aux milliers de sp\u00e9cialistes de la pr\u00e9diction et de chercheurs en biologie structurale qui ont, depuis lors, pass\u00e9 des ann\u00e9es \u00e0 r\u00e9aliser des exp\u00e9rimentations sur les prot\u00e9ines. La base de donn\u00e9es \u00e9largit consid\u00e9rablement les connaissances accumul\u00e9es sur les structures prot\u00e9iques, en faisant plus que doubler le nombre de structures prot\u00e9iques humaines de haute pr\u00e9cision mises \u00e0 disposition des chercheurs. L&#8217;am\u00e9lioration de la compr\u00e9hension de ces \u00e9l\u00e9ments constitutifs de la vie, qui sont \u00e0 la base de tous les processus biologiques de chaque \u00eatre vivant, permettra aux chercheurs d&#8217;acc\u00e9l\u00e9rer leurs travaux dans un tr\u00e8s grand nombre de domaines.<\/p>\n\n\n\n<p>La semaine derni\u00e8re, la m\u00e9thodologie \u00e0 la base de la derni\u00e8re version hautement innovante d&#8217;AlphaFold &#8211; le sophistiqu\u00e9 syst\u00e8me d&#8217;intelligence artificielle annonc\u00e9 en d\u00e9cembre dernier qui permet ces pr\u00e9dictions de structure &#8211; et son code source ouvert (<em>open source code<\/em>), ont \u00e9t\u00e9 publi\u00e9s dans <em>Nature<\/em>. L&#8217;annonce faite aujourd&#8217;hui co\u00efncide avec la publication d&#8217;un deuxi\u00e8me article dans <em><a href=\"https:\/\/www.nature.com\/articles\/s41586-021-03828-1\">Nature<\/a><\/em> qui donne l&#8217;image la plus compl\u00e8te des prot\u00e9ines qui composent le prot\u00e9ome humain, et la publication de 20 organismes suppl\u00e9mentaires importants pour la recherche en biologie.<\/p>\n\n\n\n<p>&#8220;Notre objectif \u00e0 DeepMind a toujours \u00e9t\u00e9 de construire l&#8217;intelligence artificielle et de l&#8217;utiliser comme un outil pour aider \u00e0 acc\u00e9l\u00e9rer le rythme de la d\u00e9couverte scientifique, faisant ainsi progresser notre compr\u00e9hension du monde qui nous entoure&#8221;, a d\u00e9clar\u00e9 le fondateur et PDG de DeepMind, Demis Hassabis, PhD. &#8220;Nous avons utilis\u00e9 AlphaFold pour g\u00e9n\u00e9rer l&#8217;image la plus compl\u00e8te et la plus pr\u00e9cise du prot\u00e9ome humain. Nous pensons que cela repr\u00e9sente, \u00e0 ce jour, la contribution la plus importante de l&#8217;intelligence artificielle \u00e0 l&#8217;avancement des connaissances scientifiques, et illustre parfaitement le type d&#8217;avantages que celle-ci peut apporter \u00e0 la soci\u00e9t\u00e9.&#8221;<\/p>\n\n\n\n<h2 class=\"wp-block-heading\"><strong>AlphaFold aide d\u00e9j\u00e0 les scientifiques \u00e0 acc\u00e9l\u00e9rer les d\u00e9couvertes&nbsp;<\/strong><\/h2>\n\n\n\n<p>La possibilit\u00e9 de pr\u00e9dire la forme d&#8217;une prot\u00e9ine par ordinateur \u00e0 partir de sa s\u00e9quence d&#8217;acides amin\u00e9s &#8211; plut\u00f4t que de la d\u00e9terminer exp\u00e9rimentalement par des ann\u00e9es de techniques minutieuses, ardues et souvent co\u00fbteuses &#8211; aide d\u00e9j\u00e0 les scientifiques \u00e0 r\u00e9aliser en quelques mois ce qui prenait auparavant des ann\u00e9es.<\/p>\n\n\n\n<p>&#8220;La base de donn\u00e9es AlphaFold est un exemple parfait du cercle vertueux de la science ouverte&#8221;, a d\u00e9clar\u00e9 Edith Heard, directrice g\u00e9n\u00e9rale de l&#8217;EMBL. &#8220;AlphaFold a \u00e9t\u00e9 form\u00e9 en utilisant des donn\u00e9es provenant de ressources publiques construites par la communaut\u00e9 scientifique, il est donc logique que ces pr\u00e9dictions soient rendues publiques. Le partage ouvert et gratuit des pr\u00e9dictions d&#8217;AlphaFold permettra aux chercheurs du monde entier d&#8217;acqu\u00e9rir de nouvelles connaissances et de stimuler la d\u00e9couverte. Je pense qu&#8217;AlphaFold est v\u00e9ritablement une r\u00e9volution pour les sciences de la vie, tout comme l&#8217;a \u00e9t\u00e9 la g\u00e9nomique il y a plusieurs d\u00e9cennies, et je suis tr\u00e8s fi\u00e8re que l&#8217;EMBL ait pu aider DeepMind \u00e0 rendre publique cette ressource remarquable&#8221;.<\/p>\n\n\n\n<p>AlphaFold est d\u00e9j\u00e0 utilis\u00e9 par des partenaires tels que l&#8217;<a href=\"https:\/\/dndi.org\/\">initiative Drugs for Neglected Diseases<\/a> (DNDi), qui a fait <a href=\"https:\/\/www.wired.co.uk\/article\/deepmind-alphafold-protein-diseases\">progresser la recherche de traitements permettant de sauver des vies<\/a> pour des maladies qui touchent de mani\u00e8re disproportionn\u00e9e les r\u00e9gions les plus pauvres du monde, et le <a href=\"https:\/\/www.port.ac.uk\/research\/research-centres-and-groups\/centre-for-enzyme-innovation\">Centre for Enzyme Innovation<\/a> (CEI) qui utilise AlphaFold pour aider \u00e0 concevoir des enzymes plus rapides pouvant servir \u00e0 recycler certains des plastiques \u00e0 usage unique les plus polluants. Pour les scientifiques qui d\u00e9pendent de la d\u00e9termination exp\u00e9rimentale de la structure des prot\u00e9ines, les pr\u00e9dictions d&#8217;AlphaFold ont permis d&#8217;acc\u00e9l\u00e9rer leurs recherches. Par exemple, une \u00e9quipe de l&#8217;universit\u00e9 du Colorado \u00e0 Boulder trouve prometteuse l&#8217;utilisation des pr\u00e9dictions d&#8217;AlphaFold pour \u00e9tudier la r\u00e9sistance aux antibiotiques, tandis qu&#8217;un groupe de l&#8217;universit\u00e9 de Californie \u00e0 San Francisco les a utilis\u00e9es pour <a href=\"https:\/\/www.biorxiv.org\/content\/10.1101\/2021.05.10.443524v1\">mieux comprendre la biologie du SARS-CoV-2.<\/a><\/p>\n\n\n\n<h2 class=\"wp-block-heading\"><strong>La base de donn\u00e9es de structure prot\u00e9ique AlphaFold<\/strong><\/h2>\n\n\n\n<p>La <a href=\"http:\/\/www.alphafold.ebi.ac.uk\/\">base de donn\u00e9es sur la structure des prot\u00e9ines AlphaFold<\/a> s&#8217;appuie sur de nombreuses contributions de la communaut\u00e9 scientifique internationale, ainsi que sur les innovations algorithmiques sophistiqu\u00e9es d&#8217;AlphaFold et sur les d\u00e9cennies d&#8217;exp\u00e9rience de l&#8217;EMBL-EBI en mati\u00e8re de partage des donn\u00e9es biologiques au niveau mondial. DeepMind et l&#8217;Institut europ\u00e9en de bioinformatique de l&#8217;EMBL (EMBL-EBI) donnent acc\u00e8s aux pr\u00e9dictions d&#8217;AlphaFold afin que d&#8217;autres puissent utiliser ce syst\u00e8me comme un outil permettant et acc\u00e9l\u00e9rant la recherche, et ouvrant de toutes nouvelles voies \u00e0 la d\u00e9couverte scientifique.<\/p>\n\n\n\n<p>&#8220;Ce sera l&#8217;un des ensembles de donn\u00e9es les plus importants depuis la cartographie du g\u00e9nome humain&#8221;, a d\u00e9clar\u00e9 Ewan Birney, directeur g\u00e9n\u00e9ral adjoint de l&#8217;EMBL et directeur de l&#8217;EMBL-EBI. &#8220;Rendre les pr\u00e9dictions d&#8217;AlphaFold accessibles \u00e0 la communaut\u00e9 scientifique internationale ouvre \u00e9norm\u00e9ment de nouvelles voies de recherche, depuis les maladies n\u00e9glig\u00e9es jusqu\u2019aux nouvelles enzymes pour la biotechnologie, et pour une multitude d\u2019autres domaines. Il s&#8217;agit d&#8217;un nouvel outil scientifique formidable, qui compl\u00e8te les technologies existantes, et qui nous permettra de repousser les limites de notre compr\u00e9hension du monde.&#8221;<\/p>\n\n\n\n<p>Au-del\u00e0 du prot\u00e9ome humain, la base de donn\u00e9es comprend au total environ 350 000 structures de prot\u00e9ines, dont celles de 20 organismes biologiquement significatifs tels que E.coli, la mouche du vinaigre (drosophile), la souris, le poisson z\u00e8bre, le parasite du paludisme et la bact\u00e9rie de la tuberculose. Les recherches portant sur ces organismes ont fait l&#8217;objet d&#8217;innombrables articles de recherche et de nombreuses avanc\u00e9es majeures. Ces structures permettront aux chercheurs d&#8217;une grande vari\u00e9t\u00e9 de domaines &#8211; des neurosciences \u00e0 la m\u00e9decine &#8211; d&#8217;acc\u00e9l\u00e9rer leurs travaux.<\/p>\n\n\n\n<h2 class=\"wp-block-heading\"><strong>L&#8217;avenir d&#8217;AlphaFold<\/strong><\/h2>\n\n\n\n<p>La base de donn\u00e9es et le syst\u00e8me seront p\u00e9riodiquement mis \u00e0 jour, au fur et \u00e0 mesure que nous continuons \u00e0 investir dans des am\u00e9liorations futures d&#8217;AlphaFold. Au cours des prochains mois, nous pr\u00e9voyons d&#8217;\u00e9tendre consid\u00e9rablement la couverture \u00e0 presque toutes les prot\u00e9ines s\u00e9quenc\u00e9es connues de la science &#8211; plus de 100 millions de structures couvrant la majeure partie de la base de donn\u00e9es de r\u00e9f\u00e9rence <a href=\"https:\/\/www.uniprot.org\/\">UniProt<\/a>.<\/p>\n\n\n\n<p>Pour plus de d\u00e9tails, il est possible de consulter les articles de <em>Nature<\/em> d\u00e9crivant notre <a href=\"https:\/\/www.nature.com\/articles\/s41586-021-03819-2\">m\u00e9thode compl\u00e8te<\/a> et le <a href=\"https:\/\/www.nature.com\/articles\/s41586-021-03828-1\">prot\u00e9ome humain<\/a>, et de lire les notes des auteurs <a href=\"https:\/\/deepmind.com\/research\/publications\/enabling-high-accuracy-protein-structure-prediction-at-the-proteome-scale\">ici<\/a>. Il est \u00e9galement possible de consulter le <a href=\"https:\/\/github.com\/deepmind\/alphafold\">code open source d&#8217;AlphaFold<\/a><strong> <\/strong>pour voir le fonctionnement du syst\u00e8me, et le <a href=\"https:\/\/colab.research.google.com\/github\/deepmind\/alphafold\/blob\/main\/notebooks\/AlphaFold.ipynb\">Colab notebook<\/a> pour ex\u00e9cuter des s\u00e9quences individuelles. Pour explorer les structures, il est possible de visiter la <a href=\"https:\/\/alphafold.ebi.ac.uk\/\">base de donn\u00e9es consultable<\/a> de l&#8217;EMBL-EBI, qui est ouverte et gratuite pour tous.<\/p>\n\n\n<hr class=\"vf-divider\"\/>\n\n\n<h1 class=\"wp-block-heading\" id=\"a3\"><strong>DeepMind e EMBL rendono pubblico il pi\u00f9 completo database per la predizione della struttura 3D delle proteine umane<\/strong><\/h1>\n\n\n\n\n\n\n\n<p>DeepMind ha annunciato oggi la sua partnership con il Laboratorio Europeo di Biologia Molecolare (EMBL), il principale laboratorio europeo per le scienze della vita, per il lancio del database pi\u00f9 completo e accurato per la predizione delle strutture del proteoma umano. Questi dati &#8211; comprese le strutture delle circa 20.000 proteine espresse dal genoma umano \u2013 saranno disponibili tramite accesso aperto e libero alla comunit\u00e0 scientifica. Il database e il sistema di intelligenza artificiale forniscono ai biologi strutturali nuovi potenti strumenti per esaminare la struttura tridimensionale delle proteine, e offrono dati preziosi che potrebbero accelerare i progressi futuri e annunciare una nuova era per la biologia supportata dall&#8217;intelligenza artificiale (IA).<\/p>\n\n\n\n<p>Lo scorso dicembre, Il <a href=\"https:\/\/deepmind.com\/blog\/article\/alphafold-a-solution-to-a-50-year-old-grand-challenge-in-biology\">riconoscimento<\/a> di AlphaFold da parte degli organizzatori dell\u2019iniziativa \u201c<em>Critical Assessment of protein Structure Prediction (CASP)<\/em>\u201d come soluzione alla grande sfida che risale a 50 anni fa della <a href=\"https:\/\/en.wikipedia.org\/wiki\/Protein_structure_prediction#:~:text=Protein%20structure%20prediction%20(also%20called,structure%20from%20its%20primary%20structure.\">predizione della struttura delle proteine<\/a> \u00e8 stata una svolta sorprendente nel settore. <em>L&#8217;AlphaFold Protein Structure Database<\/em> si basa su questa innovazione e sulle scoperte di generazioni di scienziati, dai primi pionieri dell&#8217;imaging e della cristallografia, alle migliaia di specialisti che hanno passato anni a studiare le proteine. Il database aumenta moltissimo la conoscenza acquisita finora, pi\u00f9 che raddoppiando il numero di strutture proteiche umane predette con grande precisione e ora messe a disposizione dei ricercatori di tutto il mondo. &nbsp;Comprendere pi\u00f9 in dettaglio come funzionano questi elementi costitutivi della vita, alla base di ogni processo biologico negli esseri viventi, permetter\u00e0 ai ricercatori di velocizzare il loro lavoro in moltissimi campi.<\/p>\n\n\n\n<p>La scorsa settimana, la metodologia alla base dell&#8217;ultima versione altamente innovativa di AlphaFold &#8211; il sofisticato sistema di IA annunciato lo scorso dicembre che guida le previsioni di struttura &#8211; e il suo codice <em>open source <\/em>sono stati pubblicati sulla rivista <em>Nature<\/em>. L&#8217;annuncio di oggi coincide con un secondo articolo su <em><a href=\"https:\/\/www.nature.com\/articles\/s41586-021-03828-1\">Nature<\/a><\/em> che fornisce il quadro pi\u00f9 completo delle proteine che compongono il proteoma umano, e la pubblicazione delle predizioni delle strutture proteiche di altri 20 organismi importanti per la ricerca biologica.<\/p>\n\n\n\n<p>&#8220;Il nostro obiettivo a DeepMind \u00e8 sempre stato quello di costruire IA e poi usarla come strumento per contribuire ad accelerare il passo della ricerca scientifica, facendo cos\u00ec progredire la nostra comprensione del mondo che ci circonda&#8221;, ha detto il fondatore e CEO di DeepMind Demis Hassabis &#8220;Abbiamo usato AlphaFold per generare il quadro pi\u00f9 completo e accurato possibile del proteoma umano. Crediamo che questo rappresenti il contributo pi\u00f9 significativo che l&#8217;IA abbia dato al progresso della conoscenza scientifica fino ad oggi, ed \u00e8 un grande esempio dei benefici che l&#8217;IA pu\u00f2 portare alla societ\u00e0&#8221;.<\/p>\n\n\n\n<h2 class=\"wp-block-heading\"><strong>AlphaFold sta gi\u00e0 aiutando gli scienziati ad accelerare la ricerca&nbsp;<\/strong><\/h2>\n\n\n\n<p>La capacit\u00e0 di prevedere la forma di una proteina tramite calcoli computazionali partendo dalla sua sequenza di aminoacidi &#8211; piuttosto che determinarla sperimentalmente attraverso anni di tecniche minuziose, laboriose e spesso costose &#8211; sta gi\u00e0 aiutando gli scienziati a raggiungere in mesi ci\u00f2 che prima richiedeva anni.<\/p>\n\n\n\n<p>&#8220;Il database AlphaFold \u00e8 un perfetto esempio del circolo virtuoso alimentato dalla scienza aperta&#8221;, ha detto la direttrice generale dell&#8217;EMBL Edith Heard. &#8220;AlphaFold \u00e8 stato istruito usando dati provenienti da basi di dati pubbliche costruite dalla comunit\u00e0 scientifica, quindi ha senso che anche &nbsp;le sue previsioni siano pubbliche. Condividere le previsioni di AlphaFold apertamente e liberamente permetter\u00e0 ai ricercatori di tutto il mondo di ottenere nuove informazioni e di avanzare rapidamente nelle loro scoperte. Credo che AlphaFold sia davvero una rivoluzione per le scienze della vita, proprio come lo \u00e8 stata la genomica diversi decenni fa, e sono molto orgogliosa che l\u2019EMBL sia stato in grado di aiutare DeepMind a garantire l&#8217;accesso aperto a questa straordinaria risorsa&#8221;.<\/p>\n\n\n\n<p>AlphaFold \u00e8 gi\u00e0 utilizzato da partner come la <a href=\"https:\/\/dndi.org\/\">Drugs for Neglected Diseases Initiative (DNDi),<\/a> per <a href=\"https:\/\/www.wired.co.uk\/article\/deepmind-alphafold-protein-diseases\">avanzare la ricerca di cure salvavita<\/a> per malattie che colpiscono in modo sproporzionato le parti pi\u00f9 povere del mondo; inoltre, il <a href=\"https:\/\/www.port.ac.uk\/research\/research-centres-and-groups\/centre-for-enzyme-innovation\">Centre for Enzyme Innovation (CEI<\/a>) sta utilizzando AlphaFold per aiutare a progettare enzimi pi\u00f9 veloci per il riciclaggio di alcune delle nostre plastiche monouso pi\u00f9 inquinanti. Le previsioni di AlphaFold stanno inoltre contribuendo ad agilizzare la ricerca degli scienziati che si affidano alla determinazione sperimentale della struttura delle proteine. Per esempio, un team dell&#8217;Universit\u00e0 del Colorado Boulder sta utilizzando le previsioni di AlphaFold per studiare la resistenza agli antibiotici, mentre un gruppo dell&#8217;Universit\u00e0 della California San Francisco le ha usate per <a href=\"https:\/\/www.biorxiv.org\/content\/10.1101\/2021.05.10.443524v1\">studiare la biologia del SARS-CoV-2<\/a>.<\/p>\n\n\n\n<h2 class=\"wp-block-heading\"><strong>Il database delle strutture proteiche AlphaFold<\/strong><\/h2>\n\n\n\n<p>L&#8217; <a href=\"https:\/\/alphafold.ebi.ac.uk\/\">AlphaFold Protein Structure Database<\/a> si basa sui numerosi contributi della comunit\u00e0 scientifica internazionale, sulle sofisticate innovazioni degli algoritmi di AlphaFold e su decenni di esperienza di EMBL-EBI nella condivisione mondiale dei dati biologici. DeepMind e l&#8217;Istituto Europeo di Bioinformatica dell&#8217;EMBL (EMBL-EBI) faciliteranno l&#8217;accesso alle previsioni di AlphaFold in modo che tutti possano utilizzare il sistema come strumento per consentire e accelerare la ricerca e aprire strade completamente nuove alla scoperta scientifica.<\/p>\n\n\n\n<p>&#8220;Questo sar\u00e0 uno degli insiemi di dati pi\u00f9 importanti dalla mappatura del genoma umano&#8221;, ha detto il vice direttore generale dell&#8217;EMBL e direttore dell&#8217;EMBL-EBI Ewan Birney. &#8220;Rendere le predizioni AlphaFold accessibili alla comunit\u00e0 scientifica internazionale apre moltissime nuove strade di ricerca, dalle malattie trascurate ai nuovi enzimi per la biotecnologia e molto altro. Si tratta di un nuovo grande strumento scientifico, che integra le tecnologie esistenti, e ci permetter\u00e0 di allargare i confini della nostra comprensione del mondo&#8221;.<\/p>\n\n\n\n<p>Oltre al proteoma umano, il database viene lanciato con circa 350.000 strutture proteiche tra cui quelle di 20 organismi biologicamente significativi come il batterio <em>E.coli<\/em>, il moscerino della frutta, il topo, zebrafish, il parassita della malaria e i batteri della tubercolosi. La ricerca su questi organismi \u00e8 stata oggetto di numerose pubblicazioni scientifiche e scoperte importanti. Queste strutture permetteranno ai ricercatori impegnati in una grande variet\u00e0 di campi &#8211; dalle neuroscienze alla medicina &#8211; di accelerare il loro lavoro.<\/p>\n\n\n\n<h2 class=\"wp-block-heading\"><strong>Il futuro di AlphaFold<\/strong><\/h2>\n\n\n\n<p>Il database e il sistema saranno aggiornati periodicamente man mano che continueremo a investire nei miglioramenti futuri di AlphaFold, e nei prossimi mesi abbiamo in programma di espandere notevolmente la copertura a quasi tutte le sequenze di proteine note &#8211; oltre 100 milioni di strutture che coprono la maggior parte del database di riferimento <a href=\"https:\/\/www.uniprot.org\/\">UniProt<\/a>.<\/p>\n\n\n\n<p>Per ulteriori dettagli, si prega di consultare gli articoli su <em>Nature<\/em> che descrivono il nostro <a href=\"https:\/\/www.nature.com\/articles\/s41586-021-03819-2\">metodo completo<\/a> e il <a href=\"https:\/\/www.nature.com\/articles\/s41586-021-03828-1\">proteoma umano<\/a>, e leggere le note degli autori <a href=\"https:\/\/deepmind.com\/research\/publications\/enabling-high-accuracy-protein-structure-prediction-at-the-proteome-scale\">qui<\/a>. Il <a href=\"https:\/\/github.com\/deepmind\/alphafold\">codice open-source di AlphaFold<\/a> mostra il funzionamento del sistema e il <a href=\"https:\/\/colab.research.google.com\/github\/deepmind\/alphafold\/blob\/main\/notebooks\/AlphaFold.ipynb\">Colab notebook<\/a> analizza le singole sequenze. Per esplorare le strutture, visitate il <a href=\"https:\/\/alphafold.ebi.ac.uk\/\">database di ricerca<\/a> di EMBL-EBI che \u00e8 aperto e gratuito per tutti.<\/p>\n\n\n<hr class=\"vf-divider\"\/>\n\n\n<h1 class=\"wp-block-heading\" id=\"a4\"><strong>DeepMind y el EMBL publican la base de datos m\u00e1s completa de predicciones de estructuras 3D de prote\u00ednas humanas hasta la fecha<\/strong><\/h1>\n\n\n\n\n\n\n\n<p>DeepMind ha anunciado hoy su colaboraci\u00f3n con el Laboratorio Europeo de Biolog\u00eda Molecular (EMBL), el principal laboratorio europeo en ciencias de la vida, para proporcionar de manera libre y abierta a la comunidad cient\u00edfica la base de datos de los modelos de predicciones de las estructuras del proteoma humano m\u00e1s completa y precisa hasta la fecha. Esto incluir\u00e1 alrededor de 20,000 prote\u00ednas expresadas por el genoma humano. La base de datos y el sistema de inteligencia artificial brindan a los bi\u00f3logos estructurales nuevas y poderosas herramientas para examinar la estructura tridimensional de las prote\u00ednas, y ofrecen un tesoro de datos que podr\u00eda abrir el camino a futuros avances y presagiar una nueva era para la biolog\u00eda basada en la inteligencia artificial.<\/p>\n\n\n\n<p>En diciembre de 2020, los organizadores de la evaluaci\u00f3n comparativa Critical Assessment of Protein Structure Prediction (CASP) <a href=\"https:\/\/deepmind.com\/blog\/article\/alphafold-a-solution-to-a-50-year-old-grand-challenge-in-biology\">reconocieron<\/a> AlphaFold como una soluci\u00f3n al gran desaf\u00edo de m\u00e1s de 50 a\u00f1os de <a href=\"https:\/\/es.wikipedia.org\/wiki\/Predicci%C3%B3n_de_la_estructura_de_las_prote%C3%ADnas\">predecir la estructura de prote\u00ednas<\/a>, lo que signific\u00f3 un logro asombroso en el campo. La base de datos de estructura de prote\u00ednas AlphaFold (AlphaFold Protein Structure Database) se basa en esta innovaci\u00f3n y en los descubrimientos de generaciones de cient\u00edficos y cient\u00edficas, desde los pioneros y las pioneras de la cristalograf\u00eda y el an\u00e1lisis de estructura de las prote\u00ednas , hasta los miles de especialistas en predicci\u00f3n y bi\u00f3logos y bi\u00f3logas estructurales que han pasado a\u00f1os experimentando con prote\u00ednas desde entonces y que han compartido sus resultados de forma abierta. La base de datos explota y ampl\u00eda dr\u00e1sticamente el conocimiento acumulado sobre las estructuras de prote\u00ednas, m\u00e1s que duplicando el n\u00famero de estructuras de prote\u00ednas humanas con predicciones de alta precisi\u00f3n disponibles para los investigadores. Avanzar en la comprensi\u00f3n de estos componentes b\u00e1sicos de la vida, que sustentan los procesos biol\u00f3gicos en todos los seres vivos, permitir\u00e1 a los investigadores de una gran variedad de campos acelerar su trabajo.<\/p>\n\n\n\n<p>La semana pasada se public\u00f3 en la revista <em>Nature<\/em> la metodolog\u00eda de la \u00faltima e innovadora versi\u00f3n de AlphaFold, el sofisticado sistema de inteligencia artificial anunciado en diciembre pasado que impulsa estas predicciones de estructura, y su c\u00f3digo fuente abierto. El anuncio de hoy coincide con un segundo art\u00edculo de <em><a href=\"https:\/\/www.nature.com\/articles\/s41586-021-03828-1\">Nature<\/a><\/em> que proporciona la imagen m\u00e1s completa de las prote\u00ednas que componen el proteoma humano, y la publicaci\u00f3n de las prote\u00ednas de 20 organismos adicionales que son importantes para la investigaci\u00f3n biol\u00f3gica.<\/p>\n\n\n\n<p>\u201cNuestro objetivo en DeepMind siempre ha sido construir inteligencia artificial y utilizarla como una herramienta para ayudar a acelerar el ritmo del descubrimiento cient\u00edfico, y mejorar as\u00ed el conocimiento del mundo que nos rodea\u201d, dijo el fundador y director ejecutivo de DeepMind, el Dr. <strong>Demis Hassabis<\/strong>. \u201cHemos utilizado AlphaFold para generar la imagen m\u00e1s completa y precisa del proteoma humano. Creemos que esta es la contribuci\u00f3n m\u00e1s significativa que ha hecho la inteligencia artificial al avance del conocimiento cient\u00edfico hasta la fecha, y es un gran ejemplo de los tipos de beneficios que la inteligencia artificial puede aportar a la sociedad\u201d.<\/p>\n\n\n\n<h2 class=\"wp-block-heading\"><strong>AlphaFold ya est\u00e1 ayudando a los cient\u00edficos a acelerar sus descubrimientos<\/strong><\/h2>\n\n\n\n<p>La capacidad de predecir computacionalmente la forma de una prote\u00edna a partir de su secuencia de amino\u00e1cidos, en lugar de tener que determinarla experimentalmente con t\u00e9cnicas minuciosas, laboriosas, y a menudo costosas, ya est\u00e1 ayudando a los cient\u00edficos a lograr en meses lo que antes requer\u00eda a\u00f1os de trabajo.<\/p>\n\n\n\n<p>\u201cLa base de datos AlphaFold es un ejemplo perfecto del c\u00edrculo virtuoso de la ciencia abierta\u201d, dijo la directora general del EMBL, Edith Heard. \u201cAlphaFold ha sido entrenado utilizando datos de recursos p\u00fablicos creados por la comunidad cient\u00edfica, por lo que tiene sentido que sus predicciones sean p\u00fablicas. Compartir las predicciones de AlphaFold de forma abierta y gratuita permitir\u00e1 a los investigadores de todo el mundo obtener nuevos conocimientos e impulsar nuevos descubrimientos. Creo que AlphaFold es una verdadera revoluci\u00f3n para las ciencias de la vida, as\u00ed como fue la gen\u00f3mica hace varias d\u00e9cadas y estoy muy orgullosa de que el EMBL haya podido ayudar a DeepMind a permitir el acceso abierto a este recurso extraordinario&#8221;.<\/p>\n\n\n\n<p>AlphaFold ya est\u00e1 siendo utilizado por socios como la Iniciativa de Medicamentos para Enfermedades Desatendidas (<a href=\"https:\/\/dndi.org\/\">DNDi<\/a>, por sus siglas en ingl\u00e9s), que ha <a href=\"https:\/\/www.wired.co.uk\/article\/deepmind-alphafold-protein-diseases\">avanzado en su investigaci\u00f3n sobre curas que salvan vidas<\/a> para enfermedades que afectan de manera desproporcionada a las zonas m\u00e1s pobres del mundo, o el Centro de Innovaci\u00f3n Enzim\u00e1tica (<a href=\"https:\/\/www.port.ac.uk\/research\/research-centres-and-groups\/centre-for-enzyme-innovation\">CEI<\/a>) que utiliza AlphaFold para ayudar a dise\u00f1ar enzimas m\u00e1s r\u00e1pidas para reciclar algunos de los pl\u00e1sticos m\u00e1s contaminantes de un solo uso. AlphaFold ha ayudado a acelerar la investigaci\u00f3n de aquellos cient\u00edficos y cient\u00edficas que trabajan en la determinaci\u00f3n experimental de la estructura de las prote\u00ednas. Por ejemplo, un equipo de la Universidad de Colorado en Boulder utiliza las predicciones de AlphaFold para estudiar la resistencia a los antibi\u00f3ticos, mientras que un grupo de la Universidad de California en San Francisco las ha utilizado para <a href=\"https:\/\/www.biorxiv.org\/content\/10.1101\/2021.05.10.443524v1\">estudiar la biolog\u00eda del SARS-CoV-2<\/a>.<\/p>\n\n\n\n<h2 class=\"wp-block-heading\"><strong>La base de datos de estructura de prote\u00ednas AlphaFold (AlphaFold Protein Structure Database)<\/strong><\/h2>\n\n\n\n<p>La <a href=\"http:\/\/www.alphafold.ebi.ac.uk\">base de datos de estructura de prote\u00ednas AlphaFold<\/a> est\u00e1 basada en muchas contribuciones de la comunidad cient\u00edfica internacional, as\u00ed como en las refinadas innovaciones algor\u00edtmicas de AlphaFold y en las d\u00e9cadas de experiencia del Instituto Europeo de Bioinform\u00e1tica del EMBL (EMBL-EBI) compartiendo datos biol\u00f3gicos mundiales. DeepMind y el EMBL-EBI est\u00e1n dando libre acceso a las predicciones de AlphaFold para que cualquiera pueda usar el sistema con el fin de permitir y acelerar la investigaci\u00f3n y explorar nuevas v\u00edas de conocimiento cient\u00edfico.<\/p>\n\n\n\n<p>\u201cEste ser\u00e1 uno de los conjuntos de datos m\u00e1s importantes desde el mapa del Genoma Humano\u201d, ha dicho el Director General Adjunto del EMBL y el director del EMBL-EBI, Ewan Birney. \u201cHacer que las predicciones de AlphaFold sean accesibles a la comunidad cient\u00edfica internacional abre muchas nuevas v\u00edas de investigaci\u00f3n, desde enfermedades desatendidas hasta nuevas enzimas para la biotecnolog\u00eda y mucho m\u00e1s. Esta es una nueva y gran herramienta cient\u00edfica, que complementa las tecnolog\u00edas existentes y nos permitir\u00e1 ampliar los l\u00edmites de nuestra comprensi\u00f3n del mundo &#8220;.<\/p>\n\n\n\n<p>Entre las primeras m\u00e1s de 350.000 estructuras publicadas en la base de datos, adem\u00e1s del proteoma humano, est\u00e1n las prote\u00ednas de 20 organismos biol\u00f3gicamente significativos como<em> E. coli<\/em>, la mosca de la fruta, el rat\u00f3n, el pez cebra, el par\u00e1sito de la malaria y las bacterias de la tuberculosis. Se han realizado muchas investigaciones importantes sobre estos organismos, y tener estas estructuras a disposici\u00f3n permitir\u00e1 a muchos investigadores de campos muy diferentes, desde la neurociencia hasta la medicina, acelerar su trabajo.<\/p>\n\n\n\n<h2 class=\"wp-block-heading\"><strong>El futuro de AlphaFold<\/strong><\/h2>\n\n\n\n<p>La base de datos y el sistema ser\u00e1n actualizados peri\u00f3dicamente a medida que se contin\u00fae invirtiendo en mejoras futuras de AlphaFold, y en los pr\u00f3ximos meses se planea expandir enormemente la cobertura a casi todas las prote\u00ednas secuenciadas conocidas por la ciencia: m\u00e1s de 100 millones de estructuras que incluyen la mayor\u00eda de UniProt, la base de datos referencia.<\/p>\n\n\n\n<p>Para m\u00e1s detalles, se pueden consultar los art\u00edculos de <em>Nature<\/em> que describen el <a href=\"https:\/\/www.nature.com\/articles\/s41586-021-03819-2\">m\u00e9todo completo<\/a> y el <a href=\"https:\/\/www.nature.com\/articles\/s41586-021-03828-1\">proteoma humano<\/a>, y leer las notas de los autores <a href=\"https:\/\/deepmind.com\/research\/publications\/enabling-high-accuracy-protein-structure-prediction-at-the-proteome-scale\">aqu\u00ed<\/a>. Para ver el funcionamiento del sistema, se puede ver el <a href=\"https:\/\/github.com\/deepmind\/alphafold\">c\u00f3digo de fuente abierta para AlphaFold<\/a>, y el <a href=\"https:\/\/colab.research.google.com\/github\/deepmind\/alphafold\/blob\/main\/notebooks\/AlphaFold.ipynb\">cuaderno Colab<\/a> para ejecutar secuencias individuales. Para explorar las estructuras, se puede visitar la <a href=\"https:\/\/alphafold.ebi.ac.uk\/\">base de datos de b\u00fasqueda<\/a> del EMBL-EBI, abierta y gratuita para todo el mundo.<\/p>\n","protected":false},"excerpt":{"rendered":"<p>Partners use AlphaFold, the AI system recognised last year as a solution to the protein structure prediction problem, to release more than 350,000 protein structure predictions including the entire human proteome to the scientific community.<\/p>\n","protected":false},"author":77,"featured_media":41174,"comment_status":"closed","ping_status":"closed","sticky":false,"template":"","format":"standard","meta":{"_acf_changed":false,"footnotes":""},"categories":[2,17591],"tags":[12758,4718,782,5088,36,556,315,1317,188,5752],"embl_taxonomy":[2906,11950],"class_list":["post-41148","post","type-post","status-publish","format-standard","has-post-thumbnail","hentry","category-science","category-science-technology","tag-alphafold","tag-artificial-intelligence","tag-database","tag-deep-learning","tag-embl-ebi","tag-open-access","tag-open-data","tag-open-science","tag-protein-data-bank","tag-protein-structure","embl_taxonomy-embl-ebi","embl_taxonomy-open-science-at-embl"],"acf":{"featured":true,"show_featured_image":false,"article_intro":"<p>Partners use AlphaFold, the AI system recognised last year as a solution to the protein structure prediction problem, to release more than 350,000 protein structure predictions including the entire human proteome to the scientific community.<\/p>\n","related_links":[{"link_description":"Solving the protein structure puzzle","link_url":"https:\/\/www.embl.org\/news\/science\/alphafold-protein-structure\/"},{"link_description":"Accessible 3D protein models to accelerate scientific discovery ","link_url":"https:\/\/www.embl.org\/news\/science\/thornton-alphafold\/"},{"link_description":"Great expectations - the potential impacts of AlphaFold DB","link_url":"https:\/\/www.embl.org\/news\/science\/alphafold-potential-impacts\/"}],"source_article":[{"publication_title":"Highly accurate protein structure prediction for the human proteome ","publication_link":{"title":"","url":"https:\/\/www.nature.com\/articles\/s41586-021-03828-1","target":""},"publication_authors":"Tunyasuvunakool K., et al.","publication_source":"Nature","publication_date":"22 July 2021","publication_doi":"10.1038\/s41586-021-03819-2"},{"publication_title":"Highly accurate protein structure prediction with AlphaFold","publication_link":{"title":"","url":"https:\/\/www.nature.com\/articles\/s41586-021-03819-2","target":""},"publication_authors":"Jumper J., et al.","publication_source":"Nature","publication_date":"15 July 2021","publication_doi":"10.1038\/s41586-021-03819-2"}],"in_this_article":false,"press_contact":"EMBL-EBI Generic","vf_locked":false,"field_target_display":"","field_article_language":{"value":"english","label":"English"},"article_translations":false,"languages":""},"embl_taxonomy_terms":[{"uuid":"a:3:{i:0;s:36:\"b14d3f13-5670-44fb-8970-e54dfd9c921a\";i:1;s:36:\"89e00fee-87f4-482e-a801-4c3548bb6a58\";i:2;s:36:\"a99d1a7c-ca83-4c00-ab61-d082d3e41ce3\";}","parents":[],"name":["EMBL-EBI"],"slug":"embl-ebi","description":"Where &gt; All EMBL sites &gt; EMBL-EBI"},{"uuid":"a:3:{i:0;s:36:\"302cfdf7-365b-462a-be65-82c7b783ebf7\";i:1;s:36:\"bc1eaadd-1f50-4140-8e9b-e58fc33a39fc\";i:2;s:36:\"01c833a7-6d73-487f-8ef2-21cf5c82bc28\";}","parents":[],"name":["Open science at EMBL"],"slug":"open-science-at-embl","description":"What &gt; About EMBL &gt; Open science at EMBL"}],"yoast_head":"<!-- This site is optimized with the Yoast SEO plugin v26.2 - https:\/\/yoast.com\/wordpress\/plugins\/seo\/ -->\n<title>DeepMind and EMBL release the most complete database of predicted 3D structures of human proteins | EMBL<\/title>\n<meta name=\"description\" content=\"Partners use AlphaFold to release more than 350,000 protein structure predictions including the entire human proteome to the scientific community.\" \/>\n<meta name=\"robots\" content=\"index, follow, max-snippet:-1, max-image-preview:large, max-video-preview:-1\" \/>\n<link rel=\"canonical\" href=\"https:\/\/www.embl.org\/news\/science\/alphafold-database-launch\/\" \/>\n<meta property=\"og:locale\" content=\"en_US\" \/>\n<meta property=\"og:type\" content=\"article\" \/>\n<meta property=\"og:title\" content=\"DeepMind and EMBL release the most complete database of predicted 3D structures of human proteins | EMBL\" \/>\n<meta property=\"og:description\" content=\"Partners use AlphaFold to release more than 350,000 protein structure predictions including the entire human proteome to the scientific community.\" \/>\n<meta property=\"og:url\" content=\"https:\/\/www.embl.org\/news\/science\/alphafold-database-launch\/\" \/>\n<meta property=\"og:site_name\" content=\"EMBL\" \/>\n<meta property=\"article:publisher\" content=\"https:\/\/www.facebook.com\/embl.org\/\" \/>\n<meta property=\"article:published_time\" content=\"2021-07-22T15:00:20+00:00\" \/>\n<meta property=\"article:modified_time\" content=\"2024-03-22T10:22:50+00:00\" \/>\n<meta property=\"og:image\" content=\"https:\/\/www.embl.org\/news\/wp-content\/uploads\/2021\/07\/2021_AlphaFold_press_release2_Credit_Karen_Arnott_1000x600.jpg\" \/>\n\t<meta property=\"og:image:width\" content=\"1000\" \/>\n\t<meta property=\"og:image:height\" content=\"600\" \/>\n\t<meta property=\"og:image:type\" content=\"image\/jpeg\" \/>\n<meta name=\"author\" content=\"Vicky Hatch\" \/>\n<meta name=\"twitter:card\" content=\"summary_large_image\" \/>\n<meta name=\"twitter:creator\" content=\"@embl\" \/>\n<meta name=\"twitter:site\" content=\"@embl\" \/>\n<meta name=\"twitter:label1\" content=\"Written by\" \/>\n\t<meta name=\"twitter:data1\" content=\"Vicky Hatch\" \/>\n\t<meta name=\"twitter:label2\" content=\"Est. reading time\" \/>\n\t<meta name=\"twitter:data2\" content=\"26 minutes\" \/>\n<script type=\"application\/ld+json\" class=\"yoast-schema-graph\">{\"@context\":\"https:\/\/schema.org\",\"@graph\":[{\"@type\":\"NewsArticle\",\"@id\":\"https:\/\/www.embl.org\/news\/science\/alphafold-database-launch\/#article\",\"isPartOf\":{\"@id\":\"https:\/\/www.embl.org\/news\/science\/alphafold-database-launch\/\"},\"author\":{\"name\":\"Vicky Hatch\",\"@id\":\"https:\/\/www.embl.org\/news\/#\/schema\/person\/d8477ba2d7a6164b141a3872a25ee982\"},\"headline\":\"DeepMind and EMBL release the most complete database of predicted 3D structures of human proteins\",\"datePublished\":\"2021-07-22T15:00:20+00:00\",\"dateModified\":\"2024-03-22T10:22:50+00:00\",\"mainEntityOfPage\":{\"@id\":\"https:\/\/www.embl.org\/news\/science\/alphafold-database-launch\/\"},\"wordCount\":6195,\"publisher\":{\"@id\":\"https:\/\/www.embl.org\/news\/#organization\"},\"image\":{\"@id\":\"https:\/\/www.embl.org\/news\/science\/alphafold-database-launch\/#primaryimage\"},\"thumbnailUrl\":\"https:\/\/www.embl.org\/news\/wp-content\/uploads\/2021\/07\/2021_AlphaFold_press_release2_Credit_Karen_Arnott_1000x600.jpg\",\"keywords\":[\"alphafold\",\"artificial intelligence\",\"database\",\"deep learning\",\"embl-ebi\",\"open access\",\"open data\",\"open science\",\"protein data bank\",\"protein structure\"],\"articleSection\":[\"Science\",\"Science &amp; Technology\"],\"inLanguage\":\"en-US\"},{\"@type\":\"WebPage\",\"@id\":\"https:\/\/www.embl.org\/news\/science\/alphafold-database-launch\/\",\"url\":\"https:\/\/www.embl.org\/news\/science\/alphafold-database-launch\/\",\"name\":\"DeepMind and EMBL release the most complete database of predicted 3D structures of human proteins | EMBL\",\"isPartOf\":{\"@id\":\"https:\/\/www.embl.org\/news\/#website\"},\"primaryImageOfPage\":{\"@id\":\"https:\/\/www.embl.org\/news\/science\/alphafold-database-launch\/#primaryimage\"},\"image\":{\"@id\":\"https:\/\/www.embl.org\/news\/science\/alphafold-database-launch\/#primaryimage\"},\"thumbnailUrl\":\"https:\/\/www.embl.org\/news\/wp-content\/uploads\/2021\/07\/2021_AlphaFold_press_release2_Credit_Karen_Arnott_1000x600.jpg\",\"datePublished\":\"2021-07-22T15:00:20+00:00\",\"dateModified\":\"2024-03-22T10:22:50+00:00\",\"description\":\"Partners use AlphaFold to release more than 350,000 protein structure predictions including the entire human proteome to the scientific community.\",\"inLanguage\":\"en-US\",\"potentialAction\":[{\"@type\":\"ReadAction\",\"target\":[\"https:\/\/www.embl.org\/news\/science\/alphafold-database-launch\/\"]}]},{\"@type\":\"ImageObject\",\"inLanguage\":\"en-US\",\"@id\":\"https:\/\/www.embl.org\/news\/science\/alphafold-database-launch\/#primaryimage\",\"url\":\"https:\/\/www.embl.org\/news\/wp-content\/uploads\/2021\/07\/2021_AlphaFold_press_release2_Credit_Karen_Arnott_1000x600.jpg\",\"contentUrl\":\"https:\/\/www.embl.org\/news\/wp-content\/uploads\/2021\/07\/2021_AlphaFold_press_release2_Credit_Karen_Arnott_1000x600.jpg\",\"width\":1000,\"height\":600,\"caption\":\"Protein structures representing the data obtained via AlphaFold. Source image: AlphaFold. Design credit: Karen Arnott\/EMBL-EBI\"},{\"@type\":\"WebSite\",\"@id\":\"https:\/\/www.embl.org\/news\/#website\",\"url\":\"https:\/\/www.embl.org\/news\/\",\"name\":\"European Molecular Biology Laboratory News\",\"description\":\"News from the European Molecular Biology Laboratory\",\"publisher\":{\"@id\":\"https:\/\/www.embl.org\/news\/#organization\"},\"alternateName\":\"EMBL News\",\"potentialAction\":[{\"@type\":\"SearchAction\",\"target\":{\"@type\":\"EntryPoint\",\"urlTemplate\":\"https:\/\/www.embl.org\/news\/?s={search_term_string}\"},\"query-input\":{\"@type\":\"PropertyValueSpecification\",\"valueRequired\":true,\"valueName\":\"search_term_string\"}}],\"inLanguage\":\"en-US\"},{\"@type\":\"Organization\",\"@id\":\"https:\/\/www.embl.org\/news\/#organization\",\"name\":\"European Molecular Biology Laboratory\",\"alternateName\":\"EMBL\",\"url\":\"https:\/\/www.embl.org\/news\/\",\"logo\":{\"@type\":\"ImageObject\",\"inLanguage\":\"en-US\",\"@id\":\"https:\/\/www.embl.org\/news\/#\/schema\/logo\/image\/\",\"url\":\"https:\/\/www.embl.org\/news\/wp-content\/uploads\/2025\/09\/EMBL_logo_colour-1-300x144-1.png\",\"contentUrl\":\"https:\/\/www.embl.org\/news\/wp-content\/uploads\/2025\/09\/EMBL_logo_colour-1-300x144-1.png\",\"width\":300,\"height\":144,\"caption\":\"European Molecular Biology Laboratory\"},\"image\":{\"@id\":\"https:\/\/www.embl.org\/news\/#\/schema\/logo\/image\/\"},\"sameAs\":[\"https:\/\/www.facebook.com\/embl.org\/\",\"https:\/\/x.com\/embl\",\"https:\/\/www.instagram.com\/embl_org\/\",\"https:\/\/www.linkedin.com\/company\/15813\/\",\"https:\/\/www.youtube.com\/user\/emblmedia\/\"]},{\"@type\":\"Person\",\"@id\":\"https:\/\/www.embl.org\/news\/#\/schema\/person\/d8477ba2d7a6164b141a3872a25ee982\",\"name\":\"Vicky Hatch\",\"image\":{\"@type\":\"ImageObject\",\"inLanguage\":\"en-US\",\"@id\":\"https:\/\/www.embl.org\/news\/#\/schema\/person\/image\/\",\"url\":\"https:\/\/secure.gravatar.com\/avatar\/6d864d58088d9e60f42c501aa30714d303efc1ca5aed268210905409910b90d5?s=96&d=mm&r=g\",\"contentUrl\":\"https:\/\/secure.gravatar.com\/avatar\/6d864d58088d9e60f42c501aa30714d303efc1ca5aed268210905409910b90d5?s=96&d=mm&r=g\",\"caption\":\"Vicky Hatch\"},\"url\":\"https:\/\/www.embl.org\/news\/author\/vicky-hatch-2-2-2-2-2-2-2-2-2-2-2-2-2-2-2-2-2-2--2\/\"}]}<\/script>\n<!-- \/ Yoast SEO plugin. -->","yoast_head_json":{"title":"DeepMind and EMBL release the most complete database of predicted 3D structures of human proteins | EMBL","description":"Partners use AlphaFold to release more than 350,000 protein structure predictions including the entire human proteome to the scientific community.","robots":{"index":"index","follow":"follow","max-snippet":"max-snippet:-1","max-image-preview":"max-image-preview:large","max-video-preview":"max-video-preview:-1"},"canonical":"https:\/\/www.embl.org\/news\/science\/alphafold-database-launch\/","og_locale":"en_US","og_type":"article","og_title":"DeepMind and EMBL release the most complete database of predicted 3D structures of human proteins | EMBL","og_description":"Partners use AlphaFold to release more than 350,000 protein structure predictions including the entire human proteome to the scientific community.","og_url":"https:\/\/www.embl.org\/news\/science\/alphafold-database-launch\/","og_site_name":"EMBL","article_publisher":"https:\/\/www.facebook.com\/embl.org\/","article_published_time":"2021-07-22T15:00:20+00:00","article_modified_time":"2024-03-22T10:22:50+00:00","og_image":[{"width":1000,"height":600,"url":"https:\/\/www.embl.org\/news\/wp-content\/uploads\/2021\/07\/2021_AlphaFold_press_release2_Credit_Karen_Arnott_1000x600.jpg","type":"image\/jpeg"}],"author":"Vicky Hatch","twitter_card":"summary_large_image","twitter_creator":"@embl","twitter_site":"@embl","twitter_misc":{"Written by":"Vicky Hatch","Est. reading time":"26 minutes"},"schema":{"@context":"https:\/\/schema.org","@graph":[{"@type":"NewsArticle","@id":"https:\/\/www.embl.org\/news\/science\/alphafold-database-launch\/#article","isPartOf":{"@id":"https:\/\/www.embl.org\/news\/science\/alphafold-database-launch\/"},"author":{"name":"Vicky Hatch","@id":"https:\/\/www.embl.org\/news\/#\/schema\/person\/d8477ba2d7a6164b141a3872a25ee982"},"headline":"DeepMind and EMBL release the most complete database of predicted 3D structures of human proteins","datePublished":"2021-07-22T15:00:20+00:00","dateModified":"2024-03-22T10:22:50+00:00","mainEntityOfPage":{"@id":"https:\/\/www.embl.org\/news\/science\/alphafold-database-launch\/"},"wordCount":6195,"publisher":{"@id":"https:\/\/www.embl.org\/news\/#organization"},"image":{"@id":"https:\/\/www.embl.org\/news\/science\/alphafold-database-launch\/#primaryimage"},"thumbnailUrl":"https:\/\/www.embl.org\/news\/wp-content\/uploads\/2021\/07\/2021_AlphaFold_press_release2_Credit_Karen_Arnott_1000x600.jpg","keywords":["alphafold","artificial intelligence","database","deep learning","embl-ebi","open access","open data","open science","protein data bank","protein structure"],"articleSection":["Science","Science &amp; Technology"],"inLanguage":"en-US"},{"@type":"WebPage","@id":"https:\/\/www.embl.org\/news\/science\/alphafold-database-launch\/","url":"https:\/\/www.embl.org\/news\/science\/alphafold-database-launch\/","name":"DeepMind and EMBL release the most complete database of predicted 3D structures of human proteins | EMBL","isPartOf":{"@id":"https:\/\/www.embl.org\/news\/#website"},"primaryImageOfPage":{"@id":"https:\/\/www.embl.org\/news\/science\/alphafold-database-launch\/#primaryimage"},"image":{"@id":"https:\/\/www.embl.org\/news\/science\/alphafold-database-launch\/#primaryimage"},"thumbnailUrl":"https:\/\/www.embl.org\/news\/wp-content\/uploads\/2021\/07\/2021_AlphaFold_press_release2_Credit_Karen_Arnott_1000x600.jpg","datePublished":"2021-07-22T15:00:20+00:00","dateModified":"2024-03-22T10:22:50+00:00","description":"Partners use AlphaFold to release more than 350,000 protein structure predictions including the entire human proteome to the scientific community.","inLanguage":"en-US","potentialAction":[{"@type":"ReadAction","target":["https:\/\/www.embl.org\/news\/science\/alphafold-database-launch\/"]}]},{"@type":"ImageObject","inLanguage":"en-US","@id":"https:\/\/www.embl.org\/news\/science\/alphafold-database-launch\/#primaryimage","url":"https:\/\/www.embl.org\/news\/wp-content\/uploads\/2021\/07\/2021_AlphaFold_press_release2_Credit_Karen_Arnott_1000x600.jpg","contentUrl":"https:\/\/www.embl.org\/news\/wp-content\/uploads\/2021\/07\/2021_AlphaFold_press_release2_Credit_Karen_Arnott_1000x600.jpg","width":1000,"height":600,"caption":"Protein structures representing the data obtained via AlphaFold. Source image: AlphaFold. Design credit: Karen Arnott\/EMBL-EBI"},{"@type":"WebSite","@id":"https:\/\/www.embl.org\/news\/#website","url":"https:\/\/www.embl.org\/news\/","name":"European Molecular Biology Laboratory News","description":"News from the European Molecular Biology Laboratory","publisher":{"@id":"https:\/\/www.embl.org\/news\/#organization"},"alternateName":"EMBL News","potentialAction":[{"@type":"SearchAction","target":{"@type":"EntryPoint","urlTemplate":"https:\/\/www.embl.org\/news\/?s={search_term_string}"},"query-input":{"@type":"PropertyValueSpecification","valueRequired":true,"valueName":"search_term_string"}}],"inLanguage":"en-US"},{"@type":"Organization","@id":"https:\/\/www.embl.org\/news\/#organization","name":"European Molecular Biology Laboratory","alternateName":"EMBL","url":"https:\/\/www.embl.org\/news\/","logo":{"@type":"ImageObject","inLanguage":"en-US","@id":"https:\/\/www.embl.org\/news\/#\/schema\/logo\/image\/","url":"https:\/\/www.embl.org\/news\/wp-content\/uploads\/2025\/09\/EMBL_logo_colour-1-300x144-1.png","contentUrl":"https:\/\/www.embl.org\/news\/wp-content\/uploads\/2025\/09\/EMBL_logo_colour-1-300x144-1.png","width":300,"height":144,"caption":"European Molecular Biology Laboratory"},"image":{"@id":"https:\/\/www.embl.org\/news\/#\/schema\/logo\/image\/"},"sameAs":["https:\/\/www.facebook.com\/embl.org\/","https:\/\/x.com\/embl","https:\/\/www.instagram.com\/embl_org\/","https:\/\/www.linkedin.com\/company\/15813\/","https:\/\/www.youtube.com\/user\/emblmedia\/"]},{"@type":"Person","@id":"https:\/\/www.embl.org\/news\/#\/schema\/person\/d8477ba2d7a6164b141a3872a25ee982","name":"Vicky Hatch","image":{"@type":"ImageObject","inLanguage":"en-US","@id":"https:\/\/www.embl.org\/news\/#\/schema\/person\/image\/","url":"https:\/\/secure.gravatar.com\/avatar\/6d864d58088d9e60f42c501aa30714d303efc1ca5aed268210905409910b90d5?s=96&d=mm&r=g","contentUrl":"https:\/\/secure.gravatar.com\/avatar\/6d864d58088d9e60f42c501aa30714d303efc1ca5aed268210905409910b90d5?s=96&d=mm&r=g","caption":"Vicky Hatch"},"url":"https:\/\/www.embl.org\/news\/author\/vicky-hatch-2-2-2-2-2-2-2-2-2-2-2-2-2-2-2-2-2-2--2\/"}]}},"field_target_display":"","field_article_language":{"value":"english","label":"English"},"fimg_url":"https:\/\/www.embl.org\/news\/wp-content\/uploads\/2021\/07\/2021_AlphaFold_press_release2_Credit_Karen_Arnott_1000x600.jpg","featured_image_src":"https:\/\/www.embl.org\/news\/wp-content\/uploads\/2021\/07\/2021_AlphaFold_press_release2_Credit_Karen_Arnott_1000x600.jpg","_links":{"self":[{"href":"https:\/\/www.embl.org\/news\/wp-json\/wp\/v2\/posts\/41148","targetHints":{"allow":["GET"]}}],"collection":[{"href":"https:\/\/www.embl.org\/news\/wp-json\/wp\/v2\/posts"}],"about":[{"href":"https:\/\/www.embl.org\/news\/wp-json\/wp\/v2\/types\/post"}],"author":[{"embeddable":true,"href":"https:\/\/www.embl.org\/news\/wp-json\/wp\/v2\/users\/77"}],"replies":[{"embeddable":true,"href":"https:\/\/www.embl.org\/news\/wp-json\/wp\/v2\/comments?post=41148"}],"version-history":[{"count":19,"href":"https:\/\/www.embl.org\/news\/wp-json\/wp\/v2\/posts\/41148\/revisions"}],"predecessor-version":[{"id":41248,"href":"https:\/\/www.embl.org\/news\/wp-json\/wp\/v2\/posts\/41148\/revisions\/41248"}],"wp:featuredmedia":[{"embeddable":true,"href":"https:\/\/www.embl.org\/news\/wp-json\/wp\/v2\/media\/41174"}],"wp:attachment":[{"href":"https:\/\/www.embl.org\/news\/wp-json\/wp\/v2\/media?parent=41148"}],"wp:term":[{"taxonomy":"category","embeddable":true,"href":"https:\/\/www.embl.org\/news\/wp-json\/wp\/v2\/categories?post=41148"},{"taxonomy":"post_tag","embeddable":true,"href":"https:\/\/www.embl.org\/news\/wp-json\/wp\/v2\/tags?post=41148"},{"taxonomy":"embl_taxonomy","embeddable":true,"href":"https:\/\/www.embl.org\/news\/wp-json\/wp\/v2\/embl_taxonomy?post=41148"}],"curies":[{"name":"wp","href":"https:\/\/api.w.org\/{rel}","templated":true}]}}