{
  "docId": "019dd923-5e88-73ef-bd5d-06b04d219fea",
  "docSlug": "dd91c78f6570bf29",
  "documentTitle": "2023 Air Street Capital The State of AI Report 2023",
  "authorId": "AirStreetCapital",
  "authorName": "Air Street Capital",
  "documentKindSlug": "consulting-deck",
  "documentKindLabel": "Consulting deck",
  "sourceTypeSlug": "vc_research",
  "sourceTypeLabel": "VC research",
  "presentationDate": null,
  "orientation": "landscape",
  "aspectRatio": 1.777,
  "pageNumber": 59,
  "pageCount": 163,
  "prevPage": 58,
  "nextPage": 60,
  "slideType": "industry_trends",
  "function": "present_framework",
  "density": "dense",
  "nDataPoints": 0,
  "notes": "The slide explains the technical application of generative AI (diffusion models) in biotechnology, specifically protein design.",
  "elementsJson": [
    "headline_text",
    "bullet_list",
    "paragraph",
    "process_diagram"
  ],
  "metadataConfidence": 1,
  "imagePath": null,
  "slideHref": "/slides/019dd923-5e88-73ef-bd5d-06b04d219fea/59",
  "deckHref": "/decks/019dd923-5e88-73ef-bd5d-06b04d219fea",
  "deckJsonHref": "/decks/019dd923-5e88-73ef-bd5d-06b04d219fea.json",
  "deckAnchorHref": "/decks/019dd923-5e88-73ef-bd5d-06b04d219fea#slide-59",
  "components": [
    {
      "bbox": null,
      "kind": "callout",
      "text": "A model called RFdiffusion takes advantage of the high precision, residue-level resolution protein structure prediction capabilities of RoseTTAFold to fine-tune it as the denoising network in a generative diffusion model.",
      "attrs": null,
      "subkind": null,
      "toolName": "Visual emphasis",
      "toolSlug": "visual-emphasis",
      "confidence": null,
      "componentId": "019dd952-47c4-7719-a225-55720a17da42",
      "frameworkName": null,
      "frameworkSlug": null
    },
    {
      "bbox": {
        "h": 0.55,
        "w": 0.25,
        "x": 0.72,
        "y": 0.35
      },
      "kind": "image",
      "text": "Diagram showing the diffusion process from noise to structured protein designs (Unconditional, Symmetric oligomers, Binder design, Motif scaffolding, Symmetric scaffolding).",
      "attrs": null,
      "subkind": "infographic",
      "toolName": null,
      "toolSlug": null,
      "confidence": null,
      "componentId": "d84b752f-d53f-4442-be2c-5cfe64649c17",
      "frameworkName": null,
      "frameworkSlug": null
    },
    {
      "bbox": {
        "h": 0.45,
        "w": 0.65,
        "x": 0.05,
        "y": 0.38
      },
      "kind": "list",
      "text": "A model called RFdiffusion takes advantage of the high precision, residue-level resolution protein structure prediction capabilities of RoseTTAFold to fine-tune it as the denoising network in a generative diffusion model using noisy structures from the Protein Data Bank.\nSimilar to AlphaFold 2, RFdiffusion is best trained when the model conditions denoising on previous predictions between timesteps.\nRFdiffusion can generate protein backbones with desired features and ProteinMPNN can then be used to design sequences that encode these generated structures.\nThe model can produce backbone designs for protein monomers, protein binders, symmetric oligomers, enzyme active site scaffolding and more.",
      "attrs": null,
      "subkind": "bullet",
      "toolName": null,
      "toolSlug": null,
      "confidence": null,
      "componentId": "b73bca22-4a9c-4b07-bf76-f451734a0980",
      "frameworkName": null,
      "frameworkSlug": null
    },
    {
      "bbox": {
        "h": 0.12,
        "w": 0.65,
        "x": 0.05,
        "y": 0.22
      },
      "kind": "paragraph",
      "text": "Designing novel proteins from scratch such that they have desired functions or structural properties, de novo design, is of interest in both research and industry. Inspired by their success in generative modelling of images and language, diffusion models are now applied to de novo protein engineering.",
      "attrs": null,
      "subkind": "paragraph",
      "toolName": null,
      "toolSlug": null,
      "confidence": null,
      "componentId": "8725c28f-5c69-494f-81f7-b71efdaab23d",
      "frameworkName": null,
      "frameworkSlug": null
    },
    {
      "bbox": {
        "h": 0.05,
        "w": 0.9,
        "x": 0.05,
        "y": 0.14
      },
      "kind": "title",
      "text": "Diffusion models design diverse functional proteins from simple molecular specifications",
      "attrs": null,
      "subkind": "headline",
      "toolName": null,
      "toolSlug": null,
      "confidence": null,
      "componentId": "9d581af5-c71f-4df5-8346-fd631fa6383b",
      "frameworkName": null,
      "frameworkSlug": null
    }
  ],
  "metrics": [],
  "tools": [],
  "frameworks": [],
  "arcBeats": [
    {
      "to": 120,
      "from": 11,
      "beatId": "019dd95a-0682-776c-8e35-41afd44ef59f",
      "arcName": "The Triple Take",
      "arcSlug": "triple-take",
      "beatName": "The Facts (What)",
      "beatSlug": "triple-take-the-facts-what",
      "evidence": "Research + Industry sections inventory model, compute, funding facts.",
      "position": 1,
      "confidence": 78,
      "parentBeatName": "Setup",
      "parentBeatSlug": "setup"
    },
    {
      "to": 120,
      "from": 11,
      "beatId": "019dd95a-0682-776c-8e35-523bfb7f96e6",
      "arcName": "The Mountain",
      "arcSlug": "mountain",
      "beatName": "Rising Action",
      "beatSlug": "mountain-rising-action",
      "evidence": "Escalating capabilities, compute concentration and capital flows.",
      "position": 2,
      "confidence": 45,
      "parentBeatName": "Development",
      "parentBeatSlug": "development"
    }
  ],
  "loops": [
    {
      "to": 68,
      "from": 59,
      "name": "Pattern Hunter",
      "slug": "02-pattern-hunter",
      "bestFor": "Time-pressed audiences, building consensus, when data is strong",
      "matchId": "019dd95a-07fe-70ce-8d3d-f9a066ea1e78",
      "evidence": "RFdiffusion, ESMFold, PerturbNet, AlphaMissense, Med-PaLM, PLIP, CoDoC, then meta map of AI-for-science fields.",
      "position": 7,
      "objective": "AI for science accumulates breakthroughs across bio and medicine",
      "structure": "Evidence A -> Evidence B -> Evidence C -> Pattern/Conclusion",
      "confidence": 78,
      "description": "Group multiple pieces of evidence that together point to a pattern or conclusion"
    }
  ],
  "imagePathAlt": null,
  "thumbSrc": null,
  "thumbSrcAlt": null,
  "locked": true
}