Decision spine
Frames the controlling thesis, evidence status, buyer implications, and research-use limits.
Digital Health
Zemi Dossier
Research-use only
AI Biology Drug Discovery
A decision package on where AI biology is already useful, where it still fails under prospective validation, and which experiments separate model performance from decision utility.
Decision this dossier helps answer
Which AI-biology programs deserve validation spend now, and what experiment would show whether the model changes a real discovery decision?
Problem this dossier solves
Generation is no longer the scarce asset; prospective validation is.
The dossier reframes AI biology around a simple buyer question: does the model produce a decision-changing experiment, or only a plausible candidate? It maps where design systems are strong, where perturbation and translation still break, and what validation gate must come next.
What is moving: current field evidence and platform activity
What is unresolved: decision-grade proof, endpoint logic, and failure modes
What Zemi solves: evidence-gated next-study strategy
Full decision package
119-page research report plus 41-sheet Evidence & Decision Workbook
Buy now
$3,500$5,500Launch price
Controlling thesis
Generation is no longer the scarce asset; prospective validation is.
The dossier reframes AI biology around a simple buyer question: does the model produce a decision-changing experiment, or only a plausible candidate? It maps where design systems are strong, where perturbation and translation still break, and what validation gate must come next.
A Zemi Dossier turns uncertain emerging research into a decision package: what is known, what is inferred, what is hypothetical, what would falsify the thesis, and what should be built, funded, avoided, monitored, or validated next.
Research report preview
First 30 pages of the actual AI Biology Drug Discovery research report.
The public page shows the real document style, density, evidence posture, and decision framing before a buyer requests access to the full Zemi Dossier package.
Research report
The research report is the argument. It is not the whole product.
The PDF research report is the structured argument: field status, evidence maturity, buyer implications, falsifiable hypotheses, power-calculated next studies, strategic recommendations, limitations, and traceability posture.
Inspectable infrastructure
Pairs the report with claim ledgers, evidence tables, power rows, audit surfaces, and decision tools.
Executable next studies
Links hypotheses to endpoint logic, assumptions, sample-size logic, budget, and timeline planning.
Research report table of contents
Public TOC generated from the source report.
The full research report is delivered as a licensed PDF. This table shows the decision structure without publishing the full product.
| Section | What it covers | Buyer value |
|---|---|---|
| Correction Status and Evidence License | Links claims to source support so users can separate what is known, inferred, and hypothetical. | Lets the buyer audit support strength before relying on the package. |
| Executive Summary | Condenses the dossier thesis, evidence posture, buyer implications, and next-study priorities. | Helps the buyer inspect the argument, evidence posture, and decision logic before relying on the package. |
| Decision Brief Read This First | Turns the report into immediate buyer triage: what matters now, what is still uncertain, and what to do next. | Helps the buyer inspect the argument, evidence posture, and decision logic before relying on the package. |
| Scope and Evidence Integrity | Defines the evidence window, source boundaries, research-use limits, and labeling discipline. | Lets the buyer audit support strength before relying on the package. |
| Denominator definitions (every count reconciles across PDF and workbook) | Defines the populations, measurements, or comparison bases needed to interpret evidence without overgeneralizing. | Helps the buyer inspect the argument, evidence posture, and decision logic before relying on the package. |
| Decision Snapshot Overview | Maps findings to buyer actions: build, fund, avoid, monitor, or validate next. | Helps the buyer inspect the argument, evidence posture, and decision logic before relying on the package. |
| Buyer Decision Cheat Sheet | Maps findings to buyer actions: build, fund, avoid, monitor, or validate next. | Helps the buyer inspect the argument, evidence posture, and decision logic before relying on the package. |
| Buy Build Wait Avoid | Maps findings to buyer actions: build, fund, avoid, monitor, or validate next. | Helps the buyer inspect the argument, evidence posture, and decision logic before relying on the package. |
| Buyer Decision Dashboard | Maps findings to buyer actions: build, fund, avoid, monitor, or validate next. | Helps the buyer inspect the argument, evidence posture, and decision logic before relying on the package. |
| Source Quality Dashboard | Shows how strong the source base is and where conclusions depend on indirect or thinner support. | Lets the buyer audit support strength before relying on the package. |
| Discovery to Decision Cascade | Defines the governing argument and the evidence gates that would support or falsify it. | Helps the buyer inspect the argument, evidence posture, and decision logic before relying on the package. |
| Thesis stress test — what would prove this wrong (read this before you trust the rest) | Defines the governing argument and the evidence gates that would support or falsify it. | Helps the buyer inspect the argument, evidence posture, and decision logic before relying on the package. |
| The state of play at each gate (mid-2026) | Frames the field's June 2026 state, active programs, research fronts, and timing signals. | Gives the buyer a reusable way to compare programs, risks, and evidence states. |
| The Discovery-to-Decision identity (the formalism) | Summarizes a report component that ties field evidence to the dossier's decision logic. | Helps the buyer inspect the argument, evidence posture, and decision logic before relying on the package. |
| The Credibility Cascade Gate Classifier | Defines the governing argument and the evidence gates that would support or falsify it. | Gives the buyer a reusable way to compare programs, risks, and evidence states. |
| The Experiment Selection Corollary | Summarizes a report component that ties field evidence to the dossier's decision logic. | Helps the buyer inspect the argument, evidence posture, and decision logic before relying on the package. |
| Current Signal Update | Frames the field's June 2026 state, active programs, research fronts, and timing signals. | Helps the buyer inspect the argument, evidence posture, and decision logic before relying on the package. |
| Current Signal Refresh | Frames the field's June 2026 state, active programs, research fronts, and timing signals. | Helps the buyer inspect the argument, evidence posture, and decision logic before relying on the package. |
| Cross Gate Convergence Map | Shows how separate research signals converge, diverge, or interact across domains, mechanisms, or study layers. | Gives the buyer a reusable way to compare programs, risks, and evidence states. |
| The fresh scan tightens the couplings | Shows how separate research signals converge, diverge, or interact across domains, mechanisms, or study layers. | Helps the buyer inspect the argument, evidence posture, and decision logic before relying on the package. |
| Cross Domain Synthesis | Shows how separate research signals converge, diverge, or interact across domains, mechanisms, or study layers. | Helps the buyer inspect the argument, evidence posture, and decision logic before relying on the package. |
| Why Headline Metrics Mislead | Summarizes a report component that ties field evidence to the dossier's decision logic. | Helps the buyer inspect the argument, evidence posture, and decision logic before relying on the package. |
| The Dynamic Gate Wall | Operationalizes the thesis as a reusable decision surface for comparing programs or evidence states. | Gives the buyer a reusable way to compare programs, risks, and evidence states. |
| The Validation Scale Inflection | Summarizes a report component that ties field evidence to the dossier's decision logic. | Helps the buyer inspect the argument, evidence posture, and decision logic before relying on the package. |
| Evidence Maturity and Capability Gate Matrix | Links claims to source support so users can separate what is known, inferred, and hypothetical. | Gives the buyer a reusable way to compare programs, risks, and evidence states. |
| Competitive Landscape | Frames the field's June 2026 state, active programs, research fronts, and timing signals. | Connects research claims to practical diligence and monitoring decisions. |
| What the Marquee Programs Show | Summarizes a report component that ties field evidence to the dossier's decision logic. | Helps the buyer inspect the argument, evidence posture, and decision logic before relying on the package. |
| The Economics of the Validation Bottleneck | Connects research maturity to competitive position, program strategy, and diligence watchpoints. | Helps the buyer inspect the argument, evidence posture, and decision logic before relying on the package. |
| The Three Hard Questions | Operationalizes the thesis as a reusable decision surface for comparing programs or evidence states. | Helps the buyer inspect the argument, evidence posture, and decision logic before relying on the package. |
| Key Findings | Summarizes the most decision-relevant findings and their implications for the dossier thesis. | Helps the buyer inspect the argument, evidence posture, and decision logic before relying on the package. |
| Gate 1 — Representation (cleared) | Operationalizes the thesis as a reusable decision surface for comparing programs or evidence states. | Gives the buyer a reusable way to compare programs, risks, and evidence states. |
| Gate 2 — Causal grounding (binds) | Operationalizes the thesis as a reusable decision surface for comparing programs or evidence states. | Gives the buyer a reusable way to compare programs, risks, and evidence states. |
| Gate 2 — Speed vs truth | Operationalizes the thesis as a reusable decision surface for comparing programs or evidence states. | Gives the buyer a reusable way to compare programs, risks, and evidence states. |
| Gate 3 — Experimental closure | Operationalizes the thesis as a reusable decision surface for comparing programs or evidence states. | Gives the buyer a reusable way to compare programs, risks, and evidence states. |
| Gate 3 — Autonomous closure | Operationalizes the thesis as a reusable decision surface for comparing programs or evidence states. | Gives the buyer a reusable way to compare programs, risks, and evidence states. |
| Gate 4 — Transportability (the condition) | Operationalizes the thesis as a reusable decision surface for comparing programs or evidence states. | Gives the buyer a reusable way to compare programs, risks, and evidence states. |
Evidence & Decision Workbook
The workbook is the decision engine.
A locked public preview of selected workbook-derived sheets: sheet structure, decision modules, power rows, traceability, and audit posture. Licensed buyers receive the full workbook files.
Read-only workbook preview with clipped values and no formulas. Licensed buyers receive the full Evidence & Decision Workbook with editable assumptions and working files.
Workbook inventory
Delivered workbook sheets.
Full inventories belong on dossier pages, not the homepage. Sheet names below are extracted from the actual workbook file.
| # | Workbook sheet | What it does |
|---|---|---|
| 1 | Workbook Guide | Navigation layer for using the workbook, understanding sheet roles, and following the decision package. Public preview exposes Sheet, Contents fields and clipped entries such as 1. Workbook Guide, This index + grading methodology, 2. Claim Ledger. |
| 2 | Claim Ledger | Claim-by-claim ledger showing verdicts, source support, inference status, and decision relevance. Public preview exposes claim_id, claim_text, report_section, report_page fields and clipped entries such as Finding 1 — AI-driven de novo binder., AI-driven de. |
| 3 | Evidence Table | Evidence-verification surface for checking sources, support quality, and the audit trail. Public preview exposes identifier, title, authors, journal fields and clipped entries such as DOI:10.1001/jama.2025.20834, SGLT2 Inhibitors and Kidney Outcomes., Neuen. |
| 4 | Source-Support Verdicts | Independent check on whether key claims are actually supported by cited sources. Public preview exposes identifier, title, how it is used in the dossier, source_support_verdict (6-way) fields and clipped entries such as DOI:10.1001/jama.2025.20834, SGLT2. |
| 5 | Source Quality Summary | Summary of source strength, gaps, and areas where conclusions depend on weaker evidence. Public preview exposes Counts, Audited claims, 214 fields and clipped entries such as Kept (customer-facing), Excluded, 142 (27 recovered / 21 context / 2.. |
| 6 | Capability x Gate Scorecard | Executable decision surface for classifying programs, risks, gates, or evidence states. Public preview exposes AI capability, G1 Repr., G2 Causal, G3 Closure fields and clipped entries such as De novo protein/binder design, in-vivo (assay-coupled),. |
| 7 | Program Scorecard | Executable decision surface for classifying programs, risks, gates, or evidence states. Public preview exposes platform / program, G1 Repr., G2 Causal, G3 Closure fields and clipped entries such as Isomorphic Labs, preclinical (Phase 1 by end-2026), G4. |
| 8 | AI-Capability Platform Tracker | Tracks programs, trials, sponsors, or competitive context for diligence and monitoring. Public preview exposes platform/method, type, key verified result, highest validation fields and clipped entries such as RFdiffusion (antibodies), diffusion de novo. |
| 9 | Disease-Application Tracker | Tracks programs, trials, sponsors, or competitive context for diligence and monitoring. Public preview exposes disease / area, target / readout, AI method, validation fields and clipped entries such as Idiopathic pulmonary fibrosis, TNIK inhibitor. |
| 10 | Autonomous-Discovery Tracker | Tracks programs, trials, sponsors, or competitive context for diligence and monitoring. Public preview exposes system / approach, mechanism, verified result, maturity fields and clipped entries such as BATCHIE, information-maximizing batch design, validated. |
| 11 | ARPA-H Funder Program Tracker | Tracks programs, trials, sponsors, or competitive context for diligence and monitoring. Public preview exposes program / funder, focus, status / catalyst, provenance fields and clipped entries such as ARPA-H IGoR (Intelligent Generator., mechanistic disease. |
| 12 | Company Market-Signal Tracker | Tracks biomarkers or readouts that determine whether the thesis is moving toward testable evidence. Public preview exposes The platform/company ecosystem.. fields from this sheet. |
| 13 | Trial Catalyst Tracker | Tracks programs, trials, sponsors, or competitive context for diligence and monitoring. Public preview exposes asset / trial, indication, phase / status, 2026-2028 catalyst fields and clipped entries such as Rentosertib (ISM001-055, TNIK), IPF, Phase 2a. |
| 14 | Failure-Mode Tracker | Captures failure modes and constraints that could invalidate or narrow a program. Public preview exposes failure mode, gate, evidence, mitigation fields and clipped entries such as Benchmark leakage / shortcut learning, G2, perturbation<linear. |
| 15 | Regulatory Evidence-Standards | Tracks regulatory, endpoint, reimbursement, or qualification requirements that shape testing strategy. Public preview exposes item, status, gate, source / provenance fields and clipped entries such as FDA: Considerations for Use of AI to., Draft (Jan 7. |
| 16 | Biomarker Target Map | Tracks biomarkers or readouts that determine whether the thesis is moving toward testable evidence. Public preview exposes target / biomarker, disease / context, role, validation fields and clipped entries such as TNIK, IPF, first-in-class therapeutic target. |
| 17 | Figure Source Map | Traceability surface linking claims, figures, calculations, and workbook outputs back to sources. Public preview exposes figure_id, figure_title, figure_family, section_anchor fields and clipped entries such as fig_evidence_base, Source Quality Dashboard,. |
| 18 | Figure Attempt Ledger | Figure and numeric reproduction log for validating calculations, chart inputs, or displayed values. Public preview exposes figure_id, family, data_type, status fields and clipped entries such as fig_evidence_base, Source Quality Dashboard, quantitative. |
| 19 | Hypotheses Ledger | Maps hypotheses to evidence state, falsification logic, endpoints, and next-study design. Public preview exposes ID, class, title, category fields and clipped entries such as H-CCG, Platform, The Credibility-Cascade Gate.. |
| 20 | Most Valuable Next Studies | Prioritizes next studies that could change the decision, including endpoints and evidence gates. Public preview exposes study, research question, design (comparator/endpoint), success / failure fields and clipped entries such as Causal-vs-efficiency. |
| 21 | Verification & Search Log | Release-discipline surface for revisions, unresolved issues, audit findings, and readiness tracking. Public preview exposes Verification — source pack, Europe PMC / PubMed E-utils /., 140/140 resolved (89 PMID,32. fields and clipped entries such as. |
| 22 | Excluded Claims | Disposition log for rejected, bounded, or quarantined claims so they do not leak into the decision thesis. Public preview exposes bucket, primary_citation, claim_text (excerpt), disposition fields and clipped entries such as structural, Not reported in. |
| 23 | Salvage Ledger | Disposition log for rejected, bounded, or quarantined claims so they do not leak into the decision thesis. Public preview exposes source element, disposition, what changed fields and clipped entries such as Executive-Summary central tension., REUSE. |
| 24 | Power Calculations | Power-calculation layer for assumptions, endpoint logic, computed sample sizes, and sensitivity checks. Public preview exposes hypothesis_id, study_id, endpoint, test_type fields and clipped entries such as H-CCG, F1, CCG binding-gate call predicts next.. |
| 25 | Competitive Landscape | Tracks programs, trials, sponsors, or competitive context for diligence and monitoring. Public preview exposes Company, Asset / platform, Modality, Gate spanned fields and clipped entries such as Isomorphic Labs (Alphabet), AlphaFold3-derived Drug Design. |
| 26 | Validation Falsification Gates | Executable decision surface for classifying programs, risks, gates, or evidence states. Public preview exposes G-G Autonomy x trust (N3/H5), N3; H5, A prespecified human-confirmation., Unprotocolled autonomy matches. fields and clipped entries such as G-H. |
| 27 | Score Basis | Explains how scores or ratings were assigned so the decision instrument can be inspected. Public preview exposes Buyer dashboard., Low / Medium / High, Expert ordinal; actionability=how. fields and clipped entries such as Competitive diligence risk / moat.,. |
| 28 | Traceability Certificate | Traceability surface linking claims, figures, calculations, and workbook outputs back to sources. Public preview exposes claim_id, report_section, report_page, claim_text fields and clipped entries such as Finding 1 — AI-driven de novo binder., Key Findings. |
| 29 | Version Registry | Release-discipline surface for revisions, unresolved issues, audit findings, and readiness tracking. Public preview exposes v1.0, 2026-06-10, Initial Tier C build (standard), From-scratch re-grade; 214. fields and clipped entries such as v2.0, 2026-06-13,. |
| 30 | Audit Issues Log | Release-discipline surface for revisions, unresolved issues, audit findings, and readiness tracking. Public preview exposes ID, Issue, Status, Resolution / note fields and clipped entries such as AIL-01, Power Calculations sheet absent., RESOLVED. |
| 31 | Current-Signal Web Sources | Tracks biomarkers or readouts that determine whether the thesis is moving toward testable evidence. Public preview exposes slug, label (inline), date, provenance fields and clipped entries such as hitconsultant-q1-2026, HIT Consultant, Apr 2026, 2026-04-20. |
| 32 | CCG_Input_Template | Executable decision surface for classifying programs, risks, gates, or evidence states. Public preview exposes G1-REPRESENTATION-LIMITED, Representation deficit: model., ____ fields and clipped entries such as G2-CAUSAL-GROUNDING-LIMITED, Causal-grounding. |
| 33 | CCG_Feature_Score | Executable decision surface for classifying programs, risks, gates, or evidence states. Public preview exposes Gate (output class), Rubric (3 = severe deficit), Deficit 0-3 (ENTER), Weight (expert-init) fields and clipped entries such as. |
| 34 | CCG_Class_Output | Executable decision surface for classifying programs, risks, gates, or evidence states. Public preview exposes Archetype (deficits G1..G5), Binding gate, OUTPUT CLASS, 2nd-highest class fields and clipped entries such as AF3/Boltz structure model used for.,. |
| 35 | CCG_Decision_Rules | Executable decision surface for classifying programs, risks, gates, or evidence states. Public preview exposes Class (binding gate), Recommended strategy, Primary functional endpoint, Required safety readout fields and clipped entries such as. |
| 36 | CCG_Evidence_Rules | Executable decision surface for classifying programs, risks, gates, or evidence states. Public preview exposes human-prospective, 5, clinical-commit (continue., an efficacy verdict / regulatory. fields and clipped entries such as human, clinical-commit. |
| 37 | CCG_Decision_Levels | Executable decision surface for classifying programs, risks, gates, or evidence states. Public preview exposes efficacy-claim, 6 fields and clipped entries such as efficacy, works. |
| 38 | CCG_Minimum_Assay_Panel | Executable decision surface for classifying programs, risks, gates, or evidence states. Public preview exposes Gate, Minimum evidence / assay, Tier fields and clipped entries such as G1 Representation, Held-out distribution-shifted., Required. |
| 39 | CCG_Archetype_Cards | Executable decision surface for classifying programs, risks, gates, or evidence states. Public preview exposes Program / capability, Deficits G1..G5, Validation type, Intended decision fields and clipped entries such as AF3/Boltz structure model used for.,. |
| 40 | CCG_Validation_Log | Executable decision surface for classifying programs, risks, gates, or evidence states. Public preview exposes Functional endpoint, Milestone clearance = next. fields and clipped entries such as What would falsify the classifier, Binding-gate calls do not. |
| 41 | CCG_Traceability_Map | Traceability surface linking claims, figures, calculations, and workbook outputs back to sources. Public preview exposes Gate / branch, Evidence anchor, Source IDs, Thesis mapping fields and clipped entries such as G1 Representation, Evo 2 / AlphaGenome /. |
Power-calculated next studies
Power calculations are the execution bridge.
Power calculations are planning-grade research calculations for endpoint logic, assumptions, sample size, budget, and timeline. They do not prove hypotheses or constitute clinical protocol.
| Hypothesis | Primary endpoint | Test | Effect | α | Power | Computed n | Design n |
|---|---|---|---|---|---|---|---|
| H1 | Validated transportable leads per dollar (a COUNT/RATE endpoint): cau... | Two-proportion | 0.55 | 0.05 | 0.80 | 96 | 110 |
| H3 | Validated leads per fixed budget (COUNT) + time-to-decision (TIME-TO-... | Two-proportion | 0.45 | 0.05 | 0.80 | 89 | 100 |
| H6 | In-vivo therapeutic-index success: de-novo design coupled to a mechan... | Two-proportion | 0.6 | 0.05 | 0.80 | 42 | 52 |
| H4 | Phase-II success rate: early-discovery compression that CLEARS Gate-4... | Two-proportion | 0.3 | 0.05 | 0.80 | 121 | 140 |
| H2 | Prospective out-of-domain success of models selected by mid- vs top-r... | Two-proportion | 0.45 | 0.05 | 0.80 | 163 | 190 |
Release audit discipline
Audit posture is part of the product.
Public pages do not publish exact audit scores. They should state the release discipline, revision posture, and limitations without implying peer review or validated clinical guidance.
Research-use decision package
Not medical advice, clinical guidance, investment advice, or a clinical protocol.
Known / inferred / hypothetical
The package separates cited evidence from inference and frontier hypothesis generation.
Audit-gated and revision-controlled
Adversarial audit checks source support, formulas, figures, limitations, and release readiness.
Buy this dossier
Full decision package — 119-page report + 41-sheet workbook
Buy now
- Decision research report (119 pages, PDF)
- Evidence & Decision Workbook (41 sheets, .xlsx) — claims, source-support verdicts, power calculations, traceability, and audit log
- Live, editable power-calculated next studies
$3,500$5,500Launch price