Thanks to AI, we just got stunningly powerful tools to decode life.<\/p>\n\n\n\n
In two back<\/a>-to-back<\/a> papers last week, scientists at DeepMind and the University of Washington described deep learning-based methods to solve protein folding\u2014the last step of executing the programming in our DNA, and a \u201conce in a generation advance<\/a>.\u201d<\/p>\n\n\n\n Proteins are the minions of life. They form our bodies, fuel our metabolism, and are the target of most of today\u2019s medicine. They start out as a simple ribbon, translated from DNA, and subsequently fold into intricate three-dimensional architectures. Similar to Transformers, many protein units further assemble into massive, moving complexes that change their structure depending on their functional needs at the moment.<\/p>\n\n\n\n Misfolded proteins can be devastating, causing health problems from sickle cell anemia<\/a> to cancer and Alzheimer\u2019s<\/a> disease. One of biology\u2019s grandest challenges for the past 50 years has been deciphering how a simple one-dimensional ribbon-like structure turns into 3D shapes, equipped with canyons, ridges, valleys, and caves. It\u2019s as if an alien is reading the coordinates of hundreds of locations on a map of the Grand Canyon on a notebook, and reconstructing it into a 3D hologram of the actual thing\u2014without ever laying eyes on it or knowing what it should look like.<\/p>\n\n\n\n Yeah. It\u2019s hard. \u201cLots of people have broken their head on it,\u201d said<\/a> Dr. John Moult at the University of Maryland.<\/p>\n\n\n\n It\u2019s not just an academic exercise. Solving the human genome paved the way for gene therapy, CAR-T cancer<\/a> breakthroughs, and the infamous CRISPR<\/a> gene editing tool. Deciphering protein folding is bound to illuminate an entire new landscape of biology we haven\u2019t been able to study or manipulate. The fast and furious development of Covid-19 vaccines relied on scientists parsing multiple protein targets on the virus, including the spike proteins that vaccines target. Many proteins that lead to cancer have so far been out of the reach of drugs because their structure is hard to pin down.<\/p>\n\n\n\n With these new AI<\/a> tools, scientists could solve haunting medical mysteries while preparing to tackle those yet unknown. It sets the stage for better understanding our biology, informing new medicines, and even inspiring synthetic biology down the line.<\/p>\n\n\n\n \u201cWhat the DeepMind team has managed to achieve is fantastic and will change the future of structural biology and protein research,\u201d said Dr. Janet Thornton, director emeritus of the European Bioinformatics Institute.<\/p>\n\n\n\n \u201cI never thought I\u2019d see this in my lifetime,\u201d added Moult.<\/p>\n\n\n\n Picture life as a video game. If DNA is the background base code, then proteins are its execution\u2014the actual game that you play. Any bugs in DNA could trigger a crash in the program, but they could also be benign and allow the game to run as usual. In other words, most modern medicine, like gamers, cares only about the final gameplay\u2014the proteins\u2014rather than the source code that leads to it, unless something goes wrong. From diabetes medication to anti-depressants and potentially life-extending senolytics<\/a>, these drugs all work by grabbing onto proteins rather than DNA.<\/p>\n\n\n\n It\u2019s why deciphering protein structure is so important: like a key to a lock, a drug can only dock onto a protein at specific spots. Similarly, proteins often tag-team by binding together into a complex to run your body\u2019s functions\u2014say, forming a memory<\/a> or triggering an immune attack against a virus.<\/p>\n\n\n\n Proteins are made of building blocks called amino acids, which are in turn programmed by DNA. Similar to the Rosetta stone, our cells can easily translate DNA code into protein building blocks inside a clam-shell-like structure, which spits out a string of one-dimensional amino acids. These ribbons are then shuffled through a whole cellular infrastructure that allows the protein to fold into its final structure.<\/p>\n\n\n\n Back in the 1970s, the Nobel Prize winner Dr. Christian Anfinsen famously asserted that the one-dimensional sequence itself can computationally predict a protein\u2019s 3D structure. The problem is time and power: like trying to hack a password with hundreds of characters suspended in 3D space, the potential solutions are astronomical.<\/p>\n\n\n\n But we now have a tool that beats humans at finding patterns: machine learning.<\/p>\n\n\n\n In 2020, DeepMind shocked the entire field<\/a> with its entry into a legacy biennial competition. Dubbed CASP (Critical Assessment of Protein Structure Prediction), the decades-long test uses traditional lab methods for determining protein structure as its baseline to judge prediction algorithms.<\/p>\n\n\n\n The baseline\u2019s hard to get. It relies on laborious experimental techniques that can take months or even years. These methods often \u201cfreeze\u201d a protein and map its internal structure down to the atomic level using X-rays. Many proteins can\u2019t be treated this way without losing their natural structure, but the method is the best we currently have. Predictions are then compared to this gold standard to judge the underlying algorithm.<\/p>\n\n\n\n Last year DeepMind stunned everyone with their AI, blowing other competition out of the water. At the time, they were a tease, revealing little detail about their \u201cincredibly exciting<\/a>\u201d method that matched experimental results in accuracy. But the 30-minute presentation inspired Dr. Minkyung Baek at the University of Washington to develop her own approach.<\/p>\n\n\n\n Baek used a similar deep learning strategy, outlined in<\/a> a paper in Science<\/em> this week. The tool, RoseTTAFold, simultaneously considers three levels of patterns. The first looks at the amino acid building blocks of a protein and compares them to all the other sequences in a protein database.<\/p>\n\n\n\n The tool next examines how one protein\u2019s amino acids interact with another within the same protein, for example, by examining the distance between two distant building blocks. It\u2019s like looking at your hands and feet fully stretched out versus in a backbend, and measuring the distance between those extremities as you \u201cfold\u201d into a yoga pose.<\/p>\n\n\n\n Finally, the third track looks at the 3D coordinates of each atom that makes up a protein building block\u2014kind of like mapping the studs on a Lego block\u2014to compile the final 3D structure. The network then bounces back and forth between these tracks, so that one output can update another track.<\/p>\n\n\n\n The end results came close to those of DeepMind\u2019s tool, AlphaFold2, which matched the gold standard of structures obtained from experiments. Although RoseTTAFold wasn\u2019t as accurate as AlphaFold2, it seemingly required much less time and energy. For a simple protein, the algorithm was able to solve the structure using a gaming computer in about 10 minutes.<\/p>\n\n\n\n RoseTTAFold was also able to tackle the \u201cprotein assemble\u201d problem, in that it could predict the structure of proteins, made up of multiple units, by simply looking at the amino acid sequence alone. For example, they were able to predict how the structure of an immune molecule locks onto its target. Many biological functions rely on these handshakes between proteins. Being able to predict them using an algorithm opens the door to manipulating biological processes\u2014immune system, stroke, cancer, brain function\u2014that we previously couldn\u2019t access.<\/p>\n\n\n\n Since RoseTTAFold\u2019s public release in July, it\u2019s been downloaded hundreds of times, allowing other researchers to answer their baffling protein sequence questions, potentially saving years of work while collectively improving on the algorithm.<\/p>\n\n\n\n \u201cWhen there\u2019s a breakthrough like this, two years later, everyone is doing it as well if not better than before,\u201d said Moult.<\/p>\n\n\n\n Meanwhile, DeepMind is also releasing their AlphaFold2 code\u2014the one that inspired Baek.<\/p>\n\n\n\n In a new paper<\/a> in Nature<\/em>, the DeepMind team described their approach<\/a> to the 50-year mystery. The crux was to integrate multiple sources of information\u2014the evolution of a protein and its physical and geometric constraints\u2014to build a two-step system that maps out a given protein with stunningly high accuracy.<\/p>\n\n\n\n First presented at the CASP meeting, Dr. Demis Hassabis, founder and CEO of DeepMind, is ready to share the code with the world. \u201cWe pledged to share our methods and provide broad, free access to the scientific community. Today we take the first step towards delivering on that commitment by sharing AlphaFold\u2019s open-source code and publishing the system\u2019s full methodology,\u201d he wrote, adding that \u201cwe\u2019re excited to see what other new avenues of research this will enable for the community.\u201d<\/p>\n\n\n\n With the two studies, we\u2019re entering a new world of predicting\u2014and subsequently engineering or changing\u2014the building blocks of life. Dr. Andrei Lupas, an evolutionary biologist at the Max Planck Institute for Developmental Biology, and a CASP judge,\u00a0agrees<\/a>: \u201cThis will change medicine. It will change research,\u201d he said. \u201cIt will change bioengineering. It will change everything.\u201d<\/p>\n\n\n\n Image Credit: Ian Haydon,\u00a0University of Washington Institute for Protein Design<\/a><\/em><\/p>\n\n\n\n Author:<\/strong><\/p>\n\n\n\n Shelly Xuelai Fan is a neuroscientist-turned-science writer. She completed her PhD in neuroscience at the University of British Columbia, where she developed novel treatments for neurodegeneration. While studying biological brains, she became fascinated with AI and all things biotech. Following graduation, she moved to UCSF to study blood-based factors that rejuvenate aged brains. She is the…\u00a0Learn More<\/a><\/p>\n\n\n\n Original Article<\/a><\/p>","protected":false},"excerpt":{"rendered":" Thanks to AI, we just got stunningly powerful tools to decode life. In two back-to-back papers last week, scientists at DeepMind and the University of Washington described deep learning-based methods to solve protein folding\u2014the last step of executing the programming in our DNA, and a \u201conce in a generation advance.\u201d Proteins are the minions of life. They […]\n","protected":false},"author":1,"featured_media":1812,"comment_status":"open","ping_status":"open","sticky":false,"template":"","format":"standard","meta":{"episode_type":"","audio_file":"","podmotor_file_id":"","podmotor_episode_id":"","cover_image":"","cover_image_id":"","duration":"","filesize":"","filesize_raw":"","date_recorded":"","explicit":"","block":"","footnotes":""},"categories":[13],"tags":[26,53,50,27,54,51],"series":[],"class_list":["post-1811","post","type-post","status-publish","format-standard","has-post-thumbnail","hentry","category-articulos-ingles","tag-artificial-intelligence","tag-chemistry","tag-health","tag-inteligencia-artificial-2","tag-quimica","tag-salud"],"episode_featured_image":"https:\/\/singularityumexico.com\/wp-content\/uploads\/2021\/10\/AI-generated-protein-structure.jpg","episode_player_image":"https:\/\/singularityumexico.com\/wp-content\/uploads\/2023\/05\/11711533-1673157178559-89a95be153719-4-scaled.jpg","download_link":"","player_link":"","audio_player":false,"episode_data":{"playerMode":"dark","subscribeUrls":{"apple_podcasts":{"key":"apple_podcasts","url":"","label":"Apple Podcasts","class":"apple_podcasts","icon":"apple-podcasts.png"},"stitcher":{"key":"stitcher","url":"","label":"Stitcher","class":"stitcher","icon":"stitcher.png"},"google_podcasts":{"key":"google_podcasts","url":"","label":"Google Podcasts","class":"google_podcasts","icon":"google-podcasts.png"},"spotify":{"key":"spotify","url":"","label":"Spotify","class":"spotify","icon":"spotify.png"}},"rssFeedUrl":"https:\/\/singularityumexico.com\/en\/feed\/podcast\/the-feedback-loop-by-singularity","embedCode":"Birth of a Protein<\/h3>\n\n\n\n
Enter AI<\/h3>\n\n\n\n
Hacking the Body<\/h3>\n\n\n\n
\n\n\n\nProtein Folding AI Is Making a \u2018Once in a Generation\u2019 Advance in Biology<\/a><\/blockquote>