{
  "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": 62,
  "pageCount": 163,
  "prevPage": 61,
  "nextPage": 63,
  "slideType": "case_study",
  "function": "illustrate_case",
  "density": "dense",
  "nDataPoints": 6,
  "notes": "Slide from State of AI 2023 report.",
  "elementsJson": [
    "headline_text",
    "paragraph",
    "bullet_list",
    "process_diagram",
    "pie_chart",
    "logo_grid"
  ],
  "metadataConfidence": 1,
  "imagePath": null,
  "slideHref": "/slides/019dd923-5e88-73ef-bd5d-06b04d219fea/62",
  "deckHref": "/decks/019dd923-5e88-73ef-bd5d-06b04d219fea",
  "deckJsonHref": "/decks/019dd923-5e88-73ef-bd5d-06b04d219fea.json",
  "deckAnchorHref": "/decks/019dd923-5e88-73ef-bd5d-06b04d219fea#slide-62",
  "components": [
    {
      "bbox": null,
      "kind": "callout",
      "text": "A new system, AlphaMissense, makes use of AlphaFold predictions and unsupervised protein language modeling to close this gap.",
      "attrs": null,
      "subkind": null,
      "toolName": "Visual emphasis",
      "toolSlug": "visual-emphasis",
      "confidence": null,
      "componentId": "019dd952-47c4-7719-a226-9f77045da8d3",
      "frameworkName": null,
      "frameworkSlug": null
    },
    {
      "bbox": {
        "h": 0.45,
        "w": 0.28,
        "x": 0.7,
        "y": 0.45
      },
      "kind": "image",
      "text": "Diagram showing input, AlphaMissense process (structure context, protein language modeling, training variants), and output (pathogenicity score and pie chart of results).",
      "attrs": null,
      "subkind": "infographic",
      "toolName": null,
      "toolSlug": null,
      "confidence": null,
      "componentId": "8069ac08-8a37-4947-9073-5c3da4e64565",
      "frameworkName": null,
      "frameworkSlug": null
    },
    {
      "bbox": {
        "h": 0.35,
        "w": 0.65,
        "x": 0.03,
        "y": 0.48
      },
      "kind": "list",
      "text": "The AlphaMissense system is built by: (i) training on weak labels from population frequency data, avoiding circularity by not using human annotations; (ii) incorporating an unsupervised protein language modeling task to learn amino acid distributions conditioned on sequence context; and (iii) incorporating structural context by using an AlphaFold-derived system.\nAlphaMissense is then used to predict 71M missense variants, saturating the human proteome. Of these, 32% are likely pathogenic and 57% are likely benign. Additional resources include all 216M possible single amino acid substitutions across the 19,233 canonical human proteins.",
      "attrs": null,
      "subkind": "bullet",
      "toolName": null,
      "toolSlug": null,
      "confidence": null,
      "componentId": "53815c7c-f7e7-4539-86bf-d5101ac0835d",
      "frameworkName": null,
      "frameworkSlug": null
    },
    {
      "bbox": null,
      "kind": "metric",
      "text": "Pathogenicity classification: 71M",
      "attrs": null,
      "subkind": "primary",
      "toolName": "Quantification",
      "toolSlug": "quantification",
      "confidence": null,
      "componentId": "019dd952-47c4-7719-a226-a137b9428e25",
      "frameworkName": null,
      "frameworkSlug": null
    },
    {
      "bbox": {
        "h": 0.25,
        "w": 0.65,
        "x": 0.03,
        "y": 0.22
      },
      "kind": "paragraph",
      "text": "Individual changes in amino acid sequences that result from genetic variation (“missense variants”) can either be benign or result in downstream problems for protein folding, activity or stability. Over 4M of these missense variants have been identified through human population-level genome sequencing experiments. However, 98% of these variants lack any confirmed clinical classification (benign/pathogenic). A new system, AlphaMissense, makes use of AlphaFold predictions and unsupervised protein language modeling to close this gap.",
      "attrs": null,
      "subkind": "paragraph",
      "toolName": null,
      "toolSlug": null,
      "confidence": null,
      "componentId": "dc807244-71e8-4215-8e3c-72773c23eded",
      "frameworkName": null,
      "frameworkSlug": null
    },
    {
      "bbox": {
        "h": 0.05,
        "w": 0.6,
        "x": 0.03,
        "y": 0.14
      },
      "kind": "title",
      "text": "Pathogenic or not? Predicting the outcome of all single-amino acid changes",
      "attrs": null,
      "subkind": "headline",
      "toolName": null,
      "toolSlug": null,
      "confidence": null,
      "componentId": "76ee85a6-b048-4ad2-a0b3-69888d28b160",
      "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
}