{"id":526231,"date":"2026-07-14T21:21:10","date_gmt":"2026-07-14T21:21:10","guid":{"rendered":"https:\/\/www.newsbeep.com\/nz\/526231\/"},"modified":"2026-07-14T21:21:10","modified_gmt":"2026-07-14T21:21:10","slug":"protein-designs-ai-revolution-inside-david-bakers-communal-brain","status":"publish","type":"post","link":"https:\/\/www.newsbeep.com\/nz\/526231\/","title":{"rendered":"Protein Design\u2019s AI Revolution: Inside David Baker\u2019s \u201cCommunal Brain\u201d"},"content":{"rendered":"

\"185<\/a>185 former and current Baker lab members gathered at the Grand H\u00f4tel in Stockholm to celebrate the 2024 Nobel Prize in Chemistry during Nobel Week [Nobel Prize Outreach \/ Cl\u00e9ment Morin] <\/p>\n

\u201cI have this idea of a communal brain.\u201d David Baker, PhD, told me as I sat in his office at the University of Washington (UW) surrounded by colorful and complex figurines of protein structures. It was the one-year anniversary of his Nobel Prize in Chemistry win.\u00a0<\/p>\n

Just outside\u00a0his\u00a0doors,\u00a0a lab of\u00a0more than 100\u00a0researchers\u00a0was\u00a0united by the shared ambition to design proteins\u00a0from scratch (or\u00a0de novo)\u00a0for\u00a0powerful\u00a0applications\u00a0across pharmaceuticals, vaccines, biosensors, and more.\u00a0This \u201ccommunal brain\u201d housed at the UW Institute for Protein Design (IPD), where Baker led as director, was\u00a0hard at work\u00a0developing deep learning methods that could achieve atomic precision.\u00a0<\/p>\n

A small protein composed of 100 amino acids had an astronomical 20\u00b9\u2070\u2070 possible sequences. Yet, only a vanishingly tiny fraction could fold into stable, functional structures.\u00a0Misplacing a residue by an angstrom\u00a0could mean the difference between a drug binding tightly to its target or complete failure.\u00a0 \u00a0<\/p>\n

For\u00a0the\u00a0antibody\u00a0drug market worth hundreds of billions of dollars, Nathaniel Bennett, PhD, former postdoctoral researcher in the Baker lab, says AI-guided antibody design that bypasses the need for time-consuming experimental screens has long been a \u201choly grail\u201d for a breadth of indications, including cancer\u00a0and\u00a0autoimmune disease.\u00a0<\/p>\n

Last November, Bennett and colleagues published\u00a0a\u00a0Nature\u00a0paper<\/a>\u00a0demonstrating\u00a0that full length\u00a0de novo\u00a0antibodies could\u00a0bind user-specified epitopes. AI models could now construct antibody loops, the key region involved in binding that has been historically challenging to design due to its flexible nature.\u00a0\u00a0\u00a0<\/p>\n

Despite this\u00a0technological leap,\u00a0AI-designed proteins\u00a0that were manufacturable, remained stable in the body, and avoided unwanted side effects, were still a step away. The gap fueled an industry debate<\/a>\u00a0over whether\u00a0generating\u00a0de novo\u00a0medicines\u00a0was\u00a0even possible.\u00a0<\/p>\n

When I asked Baker to separate the hype from reality, he\u00a0didn\u2019t\u00a0hesitate.\u00a0<\/p>\n

\u201cThe reality is that we can now design\u00a0proteins on a computer,\u201d\u00a0Baker\u00a0explained\u00a0in our video interview<\/a>.\u00a0\u201cThe hype is that for therapeutics, there\u2019s a lot more than the basic activity of a protein binding or catalyzing a reaction. Whether de novo\u00a0proteins will revolutionize medicine will require improving our understanding of the biology.\u201d\u00a0<\/p>\n

Nobel guests\u00a0<\/p>\n

Bennett is continuing molecular design research as a co-founder at Xaira\u00a0Therapeutics.<\/a> The AI-focused biotech launched in 2024 with over $1 billion in total funding and a star-studded leadership team, including Baker, as a scientific advisor, and Marc Tessier-Lavigne,\u00a0PhD, former president of Stanford and CSO of Genentech, as CEO. Carolyn Bertozzi, PhD, Nobel laureate in chemistry, Scott Gottlieb, MD, former FDA head, and Alex Gorsky, former CEO of Johnson & Johnson,\u00a0are\u00a0among\u00a0the board of directors.\u00a0<\/p>\n

Xaira\u00a0is\u00a0among\u00a0a\u00a0staggering\u00a0list\u00a0of\u00a0biotech\u00a0companies<\/a>\u00a0that Baker has\u00a0co-founded over the past three decades.\u00a0<\/p>\n

\u201cScience all becomes obsolete quickly because the field\u2019s moving!\u201d Baker told me. \u201cThe people that you mentor are more important than any science you do. They all go on and do\u00a0great things.\u201d\u00a0<\/p>\n

2024 Nobel Week was a testament to Baker\u2019s scientific reach. Nearly 200 current and former members of his lab gathered in the Grand H\u00f4tel in Stockholm to celebrate the newly named laureate, who was among a cohort of renowned AI experts who swept the awards ceremony. \u00a0<\/p>\n

Baker shared the\u00a0Nobel Prize in Chemistry\u00a0with\u00a0Google DeepMind duo,\u00a0CEO\u00a0Demis Hassabis, PhD, and then-senior research scientist, John Jumper, PhD, whose AI model, AlphaFold, solved the protein structure prediction problem and has become one of the most widely adopted computational tools for drug discovery. \u00a0<\/p>\n

Meanwhile, the\u00a0Nobel Prize in Physics\u00a0was jointly awarded to\u00a0Geoffrey Hinton, PhD, professor emeritus at University of Toronto,\u00a0and\u00a0John Hopfield, PhD,\u00a0professor emeritus at Princeton University,\u00a0for foundational discoveries\u00a0that\u00a0enabled\u00a0machine learning\u00a0with\u00a0neural networks.\u00a0\u00a0<\/p>\n

Together, the prizes represented a pivotal moment. AI was no longer confined to computer science but had become a transformative force across disciplines, earning recognition as a breakthrough deemed to confer the \u201cgreatest benefit to humankind.\u201d\u00a0<\/p>\n

Back at the IPD, Baker\u2019s research\u00a0group\u00a0spanned multiple floors. Yet, he knew everyone\u2019s name, where they sat, and moved easily between conversations, bringing together researchers whose expertise might unlock a new direction. In the weeks\u00a0after receiving the historic Nobel call, Baker chose to remain fully present for his team, implementing a strict \u201cno travel rule,\u201d despite the avalanche of invitations and media attention that accompanied the prize.\u00a0<\/p>\n

\u201cDavid\u2019s really good at forcing you to break the ice with people,\u201d said Seth Woodbury, a\u00a0graduate student who is\u00a0designing metallohydrolases<\/a>, enzymes that cleave some of the strongest bonds in biology for sustainability applications, including degrading pollutants. \u201cOnce you talk to\u00a0your colleagues at happy hour,\u00a0it\u2019s not so scary to go ask them a question.\u201d\u00a0<\/p>\n

Woody Ahern, graduate student and co-author of the metallohydrolase\u00a0Nature\u00a0paper<\/a>, adds that Baker\u00a0has a\u00a0\u201cvery reasonable disdain for hierarchy.\u201d\u00a0<\/p>\n

\u201cAnyone can speak up in meetings. Anyone can question the work.\u00a0It breeds this culture of staying focused on what matters in an interdisciplinary way,\u201d\u00a0said Ahern.\u00a0<\/p>\n

When Ria Sonigra was applying to graduate schools in the U.S., every option felt equally far from her home in India. She recalled sending Baker a cold email with questions about the lab. He quickly replied and offered to connect her with another international student who could help her navigate the application process. Today, Sonigra is an IPD graduate student, designing programmable nanopores<\/a>\u00a0for molecular sensing and sequencing.\u00a0<\/p>\n

\u201cPeople outside the lab may think that David can\u2019t pay attention to everyone, which is not true,\u201d Sonigra said. \u201cHe knows your project and what he expects of you before the next meeting, even if he has a hundred trainees.\u201d\u00a0<\/p>\n

At one point, Baker waved me over with a smile. \u201cYou\u2019re missing chocolate hour!\u201d he\u00a0said,\u00a0inviting me\u00a0to one of many\u00a0small weekly rituals that embodied the\u00a0collaborative culture he had built.\u00a0<\/p>\n

Lowest energy\u00a0search\u00a0<\/p>\n

At\u00a0GEN\u2019s inaugural virtual event, The State of AI in Drug Discovery<\/a>,\u00a0I asked Baker for his initial reactions to winning the Nobel.\u00a0\u00a0<\/p>\n

\u201cMy group was not the first to do\u00a0protein\u00a0design,\u201d\u00a0he said humbly.\u00a0<\/p>\n

The field\u2019s early innings trace back to 1988, when William DeGrado, PhD, demonstrated that sequences not found in nature could achieve stable 3D folds<\/a>.\u00a0The\u00a0work challenged the long-held belief that functional\u00a0proteins could only arise through evolution.\u00a0<\/p>\n

Steps toward computational design\u00a0came\u00a0a decade later,\u00a0when\u00a0for the first time,\u00a0an\u00a0in silico\u00a0predicted protein was experimentally validated to adopt a target structure. The work was published in\u00a0a\u00a0Science\u00a0study<\/a>\u00a0led by Steve Mayo, PhD.\u00a0\u00a0<\/p>\n

Baker, alongside then-postdoctoral researcher, Brian Kuhlman, PhD, went a step further in 2003, expanding the design scope to include flexible backbones that represented entirely new folds<\/a>,\u00a0making it\u00a0possible\u00a0to\u00a0not only\u00a0modify\u00a0natural\u00a0proteins, but to create new ones from scratch.\u00a0<\/p>\n

\u201cThe prize was given because\u00a0protein\u00a0design has so much\u00a0promise now, and that reflects the work of the whole community,\u201d\u00a0Baker continued.\u00a0\u00a0<\/p>\n

Today, Degrado, Mayo, and Kuhlman are continuing to advance structural biology as prominent faculty members across University of California, San Francisco (UCSF), California Institute for Technology, and University of North Carolina (UNC) Chapel Hill, respectively. \u00a0<\/p>\n

\"Top7Top7 was the first protein created on a computer with a custom amino acid sequence that folds into a never-before-seen structure. When viewed at an angle, the transparent partition allows the two forms to become superimposed, illustrating the beauty of uniting sequence and structure. [UW Institute for Protein Design]Decades before\u00a0OpenAI\u00a0co-founder, Andrej Karpathy,\u00a0coined the term<\/a>\u00a0\u201cvibe coding,\u201d\u00a0Baker\u2019s team was\u00a0writing a\u00a0program in\u00a0FORTRAN.\u00a0Named\u00a0Rosetta<\/a>,\u00a0the\u00a0molecular modeling suite simulated proteins\u00a0atom-by-atom based on biophysical properties, from hydrogen bonds to backbone torsion angles.\u00a0By calculating free energy, Rosetta could estimate which sequences were most likely to achieve a desired structure: the lower the energy, the more stable the predicted fold.\u00a0<\/p>\n

Yet, a protein\u2019s energy landscape is rugged, with countless local minima among an astronomical number of conformations. Success\u00a0was\u00a0rare.\u00a0Researchers were\u00a0searching\u00a0for\u00a0a single grain of sand across the desert.\u00a0<\/p>\n

Still, \u201cRosetta was impressive,\u201d said Sierin Lim, PhD, an associate professor at Nanyang Technological University, who is among a group of researchers engineering self-assembling nanoscale containers<\/a>, known as protein\u00a0cages,\u00a0for applications across\u00a0drug discovery, imaging, and materials science.\u00a0She recalled\u00a0watching molecules move on her screen in Singapore\u00a0in the early 2000s.\u00a0At the time, Rosetta was the only program that could model proteins.\u00a0<\/p>\n

Over the next twenty years, Baker adamantly pushed Rosetta to be openly available, inviting collaborators to not only use the software, but to improve it. \u00a0<\/p>\n

PyRosetta<\/a>, a user-friendly\u00a0Python-based implementation\u00a0developed by\u00a0Johns Hopkins University researchers led by Jeffrey Gray, PhD,\u00a0broadened\u00a0Rosetta\u2019s access\u00a0for\u00a0structural biologists without a strong computational background.\u00a0Meanwhile, progress in generating high affinity and selective ligand binders<\/a>\u00a0and\u00a0epitope scaffolds for vaccine development <\/a>were bringing computational proteins closer to real-world medicines.\u00a0<\/p>\n

What started as a single lab project\u00a0grew into the Rosetta Commons, an\u00a0international collaboration spanning more than 100 laboratories.\u00a0<\/p>\n

\u201cIt was a great move making Rosetta\u00a0open, seeing what it can do now,\u201d Lim said.\u00a0\u00a0<\/p>\n

CASP14\u00a0<\/p>\n

Then came a\u00a0seminal\u00a02017\u00a0report\u00a0titled simply,\u00a0\u201cAttention Is All You Need<\/a>.\u201d\u00a0\u00a0<\/p>\n

Researchers from Google introduced the transformer, a neural network architecture that enabled machines to analyze entire sequences at once. By using a \u201cself-attention\u201d mechanism, AI models could now uncover patterns across massive datasets at unprecedented scale. Soon, large language models (LLMs)\u00a0trained on internet-scale text could not only understand, but converse in eloquent dialogue with humans.<\/p>\n

The generative AI era\u00a0had begun.\u00a0<\/p>\n

While the rest of the world\u00a0was captivated\u00a0by\u00a0chatbots,\u00a0structural biologists\u00a0were sitting on\u00a0a treasure trove of biological\u00a0data\u00a0pristine\u00a0for machine learning.\u00a0\u00a0<\/p>\n

For over fifty years, researchers had painstakingly deposited hundreds of thousands of experimentally determined structures in the Protein Data Bank (PDB) for public use. This molecular atlas now offered AI a window into the rules of biology.\u00a0<\/p>\n

In 2020,\u00a0Baker received a phone call from one of the organizers of\u00a0the Critical Assessment of protein Structure Prediction\u00a0(CASP)\u00a0competition, the biannual\u00a0experiment\u00a0that assesses the field\u2019s latest state-of-the-art models.\u00a0<\/p>\n

\u201cThe first thing he said was, \u2019David somebody has done amazingly well this year, and it isn\u2019t you!\u2019\u201d Baker recalled during his Nobel banquet speech.\u00a0\u201cThat was how\u00a0I\u00a0first learned about the work of\u00a0Demis and John.\u201d\u00a0<\/p>\n

Instead of relying on human-defined biophysical rules, AlphaFold quickly learned decades of biochemistry from the PDB, uncovering the hidden instructions governing an amino acid sequence to fold into its 3D shape. At CASP14, the model remarkably predicted structures that were indistinguishable from real-world proteins. Months of\u00a0laboratory work\u00a0turned\u00a0into a computational task completed in minutes.\u00a0<\/p>\n

Hassabis was quick to translate the breakthrough into medicine, taking the helm of DeepMind\u2019s drug discovery spinout, Isomorphic Labs, as CEO a year later.\u00a0<\/p>\n

Today, the company\u2019s IsoDD (Isomorphic Labs Drug Design Engine) platform, expands the druggable landscape by probing previously inaccessible biology, including predicting induced-fit interactions, where proteins change shape upon ligand binding, and identifying hidden binding pockets for drug targeting.\u00a0<\/p>\n

Isomorphic was betting, not on single therapeutic assets, but on a general discovery engine applicable across any disease area. That vision has since secured major pharma partnerships\u00a0with Novartis, Eli Lilly, and Johnson & Johnson.\u00a0<\/p>\n

\u201cI\u2019ve always believed the No.1 application of AI should be to improve human health,\u201d wrote Hassabis on LinkedIn when announcing\u00a0Isomorphic\u2019s\u00a0whopping\u00a0$2.1 billion\u00a0funding raise\u00a0in May.\u00a0<\/p>\n

Diffusion evolution\u00a0<\/p>\n

Concurrently,\u00a0Baker\u2019s\u00a0team\u00a0began\u00a0applying\u00a0deep\u00a0learning\u00a0to\u00a0de novo\u00a0design,\u00a0drawing inspiration from AI\u2019s emerging ability to generate realistic images. These diffusion models could operate on atomic coordinates and create entirely new protein backbones. Designs were conditioned for desired structural and functional constraints, opening the door to programmable biology.\u00a0<\/p>\n

When\u00a0Baker\u2019s team\u00a0presented\u00a0de novo design model, RFdiffusion (RoseTTAFold diffusion), in\u00a0Nature in 2023<\/a>,\u00a0Mohammed AlQuraishi, PhD,\u00a0assistant\u00a0professor of\u00a0systems\u00a0biology\u00a0at Columbia University,\u00a0described\u00a0the advance<\/a>\u00a0as \u201ca really big deal.\u201d\u00a0<\/p>\n

\u2018\u2018Prior to the \u2018diffusion evolution\u2019, the success rates were probably on the order of 1 to 10,000, if you\u2019re lucky,\u2019\u2019 AlQuraishi told me shortly after RFdiffusion\u2019s publication. \u2018\u2018With diffusion models, the success rates are closer to the single percentages when you get into the laboratory. It\u2019s a huge magnitude improvement of what it used to be.\u201d\u00a0<\/p>\n

Donald\u00a0Hilvert, PhD,\u00a0professor emeritus at\u00a0ETH\u00a0Zurich,\u00a0met Baker twenty years ago while working on\u00a0enzyme design with\u00a0Defense Advanced Research Projects Agency (DARPA). Traditional\u00a0Rosetta\u00a0methods would\u00a0carve\u00a0out\u00a0binding pockets in existing proteins\u00a0and\u00a0install a new catalytic apparatus.\u00a0<\/p>\n

\u201cBut the activities were not\u00a0very good,\u201d Hilvert\u00a0recalled.\u00a0Designing catalysis, where success depended on precisely positioning chemical groups to stabilize fleeting transition states, proved far more difficult than engineering a stable protein fold.\u00a0Rosetta struggled to achieve that level of accuracy, prompting much of the field, including Baker, to\u00a0turn\u00a0attention elsewhere.\u00a0<\/p>\n

\u201cTwo years ago, David called me and said, \u2018Why don\u2019t you come and visit? All these new AI-driven techniques are really changing the game!\u2019\u201d Hilvert told me. \u00a0<\/p>\n

Hilvert has spent the past two summers at the IPD, collaborating with Woodbury, Ahern, and IPD postdoctoral researcher, Donghyo Kim, PhD, to design metallohydrolases using RFdiffusion. He \u201chardly knew how to turn on a computer,\u201d yet was reading Python scripts and generating his first computational designs within weeks. To his amazement, experiments quickly yielded five or six promising hits.\u00a0\u00a0<\/p>\n

\u201cThere\u00a0is this\u00a0common purpose\u00a0of\u00a0people helping one another,\u201d\u00a0Hilvert\u00a0said.\u00a0\u201cDavid sets the tone from\u00a0the top.\u201d\u00a0<\/p>\n

Application generalist\u00a0<\/p>\n

As I walked through the halls of the IPD, I saw the extraordinary reach of protein design applications firsthand. Desks\u00a0were\u00a0intermingled across fields. The proximity was deliberate for ideas to travel as far as possible.\u00a0<\/p>\n

Florence Hardy, PhD, is a postdoctoral researcher tackling a new enzyme design project for global health applications, including streamlining the manufacturing process for therapeutics.\u00a0<\/p>\n

\u201cI always say that I can only think in a ten angstrom sphere at a time,\u201d she chuckled.\u00a0 \u201cThat\u2019s just as big as the active site.\u201d\u00a0<\/p>\n

\u201cMost medicines focus on inhibitors,\u201d Xinru Wang, PhD, explained when describing her postdoctoral research developing insulin agonists, or binders that lead to activation, to address metabolic disease. In contrast to blocking activity, \u201cturning on\u201d a signaling complex required precise structural tuning that was a natural\u00a0fit for the IPD\u2019s\u00a0expertise.\u00a0<\/p>\n

Last\u00a0November, Wang and colleagues\u00a0published\u00a0a study in\u00a0Molecular Cell<\/a>, demonstrating that de novo designed insulin receptor (IR) agonists could extend glucose-lowering effects. The findings offered a therapeutic alternative to escalating insulin doses, which is a known contributor to resistance. Notably, these engineered agonists avoided triggering cancer proliferation that is often associated with excessive insulin activation. Wang is currently an assistant professor at Northeastern University.\u00a0\u00a0<\/p>\n

Tabitha Tcheau designs DNA binding proteins inducible with small molecules that can recognize novel pathogens and trigger the plant immune system. The highlight of her project, she says, is the ability to span interdisciplinary subgroups, from conformational dynamics, small molecules, and nucleic acids. \u00a0<\/p>\n

\u201cOne thing that blew me away here is that people are extremely supportive,\u201d Tcheau told me. \u201cEveryone\u00a0you ask is super eager to help.\u201d\u00a0<\/p>\n

Enisha Sehgal\u00a0is\u00a0among a team of\u00a0researchers designing\u00a0sequence specific DNA binding proteins<\/a>\u00a0that can power\u00a0programmable transcription factors, targeted gene regulation, and new genome engineering\u00a0tools.\u00a0\u00a0<\/p>\n

\u201cBeing in this lab allows you to be a specialist in protein design, but a generalist in all the applications,\u201d\u00a0Sehgal\u00a0said. \u201cYou get answers faster. You can iterate faster. Science moves faster.\u201d\u00a0<\/p>\n

Visiting researcher\u00a0and machine learning\u00a0scientist,\u00a0Kieran Didi,\u00a0reiterates\u00a0how the IPD\u2019s interdisciplinary\u00a0team\u00a0enables\u00a0rapid\u00a0experimental validation of models.\u00a0\u201cI\u2019m not going to spend two months in this fantasy world of computational benchmarks,\u201d he said. \u201cIn the next week, I know if the model is actually working. Someone will quickly put it to the reality test.\u201d\u00a0<\/p>\n

Postdoctoral researcher and chemist, Declan Evans, PhD, concurs and\u00a0sees himself as the Alpha tester.\u00a0<\/p>\n

\u201cI can go straight to the developer and say, \u2018this is not how computational chemists\u00a0would use this software,\u2019\u201d Evans\u00a0said.\u00a0\u201cYou can see changes being made in real time.\u201d\u00a0<\/p>\n

Back in Baker\u2019s office, he told me about his regular weekend escape to the mountains, one of the benefits of living in Seattle. Skiing and hiking were\u00a0activities he valued highly. When asked to contribute an item to the Nobel Prize Museum, Baker chose\u00a0a\u00a0broken ski pole\u00a0as\u00a0a symbol\u00a0that progress often comes through overcoming setbacks.\u00a0<\/p>\n

\u201cBut\u00a0I don\u2019t think people\u00a0get\u00a0ideas on\u00a0top of mountains,\u201d\u00a0Baker\u00a0tempered. \u201cIf\u00a0you\u2019re\u00a0going to be a [principal investigator], you\u00a0have to\u00a0really like mentoring. For me,\u00a0it\u2019s\u00a0super fun!\u201d\u00a0<\/p>\n

Baker\u2019s most enduring creation may not be any single protein, but rather the network he built\u2014the diverse, inviting, and interconnected communal brain. \u00a0<\/p>\n","protected":false},"excerpt":{"rendered":"185 former and current Baker lab members gathered at the Grand H\u00f4tel in Stockholm to celebrate the 2024…\n","protected":false},"author":2,"featured_media":526232,"comment_status":"","ping_status":"","sticky":false,"template":"","format":"standard","meta":{"footnotes":""},"categories":[7],"tags":[363,67108,1967,111,139,69,147],"class_list":["post-526231","post","type-post","status-publish","format-standard","has-post-thumbnail","category-science","tag-artificial-intelligence","tag-gen-edge","tag-insights","tag-new-zealand","tag-newzealand","tag-nz","tag-science"],"_links":{"self":[{"href":"https:\/\/www.newsbeep.com\/nz\/wp-json\/wp\/v2\/posts\/526231","targetHints":{"allow":["GET"]}}],"collection":[{"href":"https:\/\/www.newsbeep.com\/nz\/wp-json\/wp\/v2\/posts"}],"about":[{"href":"https:\/\/www.newsbeep.com\/nz\/wp-json\/wp\/v2\/types\/post"}],"author":[{"embeddable":true,"href":"https:\/\/www.newsbeep.com\/nz\/wp-json\/wp\/v2\/users\/2"}],"replies":[{"embeddable":true,"href":"https:\/\/www.newsbeep.com\/nz\/wp-json\/wp\/v2\/comments?post=526231"}],"version-history":[{"count":0,"href":"https:\/\/www.newsbeep.com\/nz\/wp-json\/wp\/v2\/posts\/526231\/revisions"}],"wp:featuredmedia":[{"embeddable":true,"href":"https:\/\/www.newsbeep.com\/nz\/wp-json\/wp\/v2\/media\/526232"}],"wp:attachment":[{"href":"https:\/\/www.newsbeep.com\/nz\/wp-json\/wp\/v2\/media?parent=526231"}],"wp:term":[{"taxonomy":"category","embeddable":true,"href":"https:\/\/www.newsbeep.com\/nz\/wp-json\/wp\/v2\/categories?post=526231"},{"taxonomy":"post_tag","embeddable":true,"href":"https:\/\/www.newsbeep.com\/nz\/wp-json\/wp\/v2\/tags?post=526231"}],"curies":[{"name":"wp","href":"https:\/\/api.w.org\/{rel}","templated":true}]}}