DNALYZER

RESEARCH ECONOMICS / SEPTEMBER 12, 2026

What does it cost
to find out?

Compare the next decisive test with the opportunity it could unlock. Every number has an outcome, a time horizon, and an assumption you can inspect.

Planning estimates · not measured odds or approved budgets.

Probabilities are initial analyst judgments, not calibrated forecasts or an expert consensus. TAM is an annual US research-tool market scenario, not a valuation or revenue forecast. All paths share customers; their markets must not be added.

START HERE

Fund the smallest informative test.

Prioritize repeated anatomical changes and regulatory differences. Audit the atlas alongside them. Keep broad anatomy prediction paused until the representation changes.

Change the assumptions ↗

The research portfolio

Download assumptions ↓

Eight bounded workstreams plus the long-term mission. For completed pilots, budgets and probabilities refer to the next test. All work is proposed, not automatically commissioned.

01 / First priority

Repeated anatomical changes

Published organ-loss studies establish a foothold. DNALYZER has not yet produced a new association that replicates across independent transitions.

Next test · fully costed$19.3k$12.6k–$31.0k planning range · 4–6 weeks
Chance of passing this test30%15–45% judgment range · low confidence
Annual US TAM scenario$16.0M320 buyers × $50.0k/year
Test cost / annual TAM0.121%Scale comparison, not return on investment
Test, budget & assumptions

What must pass

A previously unreported association survives ancestry and sequence-quality controls and transfers to an entire evolutionary transition withheld from discovery.

Why this probability

Repeated natural changes create useful comparisons, but small event counts, missing sequence and reverse causality make novel transfer uncertain.

Stop or change direction

Stop discovery claims if too few independent transitions are usable or the held-out association fails. Recovering a known association validates the workflow, but does not pass the novelty gate.

Cost of the next check

Scientific work · 120 h × $100
$12.0k
Independent review · 24 h × $150
$3.6k
LLM allowance · 40M in / 8M out
$240
Compute, memory & storage allowance
$180
Other data costs allowance
$100
Contingency · 20%
$3.2k

Tools/data cash only: $624 including contingency, if labor is contributed. This excludes the value of human time. Actual historical spend: not metered here.

What a positive result could unlock

Independent replication and targeted functional validation: $75.0k–$250.0k additional planning allowance. No supplier quote; not included in the first test.

Potential customer and market

Comparative target-evidence service for discovery teams. Annual TAM scenario: $4.0M–$64.0M. Relevant buyers are assumed to be 5–20% of the 3,200-company US biotechnology R&D reference pool.

Annual price: $25.0k–$100.0k per organization. Neither qualification nor willingness to pay has been validated. Academic, non-US and clinical markets are excluded.

Commercial success is a separate event

20% conditional planning prior for winning three distinct paying organizations, each with a research pilot of at least $25,000, within 36 months of starting—assuming this gate passes and follow-on development and sales are funded. Joint chance under those assumptions: 6%.

This commercial prior has very low confidence and no customer-validation evidence. It is not the chance of a major biological breakthrough.

Related evidence ↗
02 / Priority test

Regulatory activity differences

Public enhancer experiments provide measured comparisons. Recovering a published example is a data audit, not a new biological experiment.

Next test · fully costed$26.4k$17.1k–$42.2k planning range · 5–8 weeks
Chance of passing this test25%10–40% judgment range · low confidence
Annual US TAM scenario$36.0M480 buyers × $75.0k/year
Test cost / annual TAM0.073%Scale comparison, not return on investment
Test, budget & assumptions

What must pass

At least 10% lower error than the strongest applicable baseline on held-out loci or studies, with the paired 95% interval excluding no improvement and no construct-family leakage.

Why this probability

Direct measurements help. Public comparison sets are limited, prior models are strong, and recognizing activity is easier than predicting a change.

Stop or change direction

Stop if comparable activity-change labels are insufficient or improvement vanishes when related loci and studies are withheld.

Cost of the next check

Scientific work · 160 h × $100
$16.0k
Independent review · 32 h × $150
$4.8k
LLM allowance · 60M in / 12M out
$360
Compute, memory & storage allowance
$650
Other data costs allowance
$150
Contingency · 20%
$4.4k

Tools/data cash only: $1.4k including contingency, if labor is contributed. This excludes the value of human time. Actual historical spend: not metered here.

What a positive result could unlock

Independent assay validation in the intended tissue context: $100.0k–$400.0k additional planning allowance. No supplier quote; not included in the first test.

Potential customer and market

Regulatory variant and enhancer evaluation for biotech R&D. Annual TAM scenario: $4.0M–$144.0M. Relevant buyers are assumed to be 5–30% of the 3,200-company US biotechnology R&D reference pool.

Annual price: $25.0k–$150.0k per organization. Neither qualification nor willingness to pay has been validated. Academic, non-US and clinical markets are excluded.

Commercial success is a separate event

25% conditional planning prior for winning three distinct paying organizations, each with a research pilot of at least $25,000, within 36 months of starting—assuming this gate passes and follow-on development and sales are funded. Joint chance under those assumptions: 6.25%.

This commercial prior has very low confidence and no customer-validation evidence. It is not the chance of a major biological breakthrough.

Related evidence ↗
03 / Foundation

Genome–phenome evidence atlas

14,329 assembly records, 3,014 species profiles and 87,265 source assertions. One complete genome downloaded and verified; coverage remains partial.

Next test · fully costed$22.6k$14.7k–$36.2k planning range · 4–6 weeks
Chance of passing this test45%25–65% judgment range · low confidence
Annual US TAM scenario$20.0M800 buyers × $25.0k/year
Test cost / annual TAM0.113%Scale comparison, not return on investment
Test, budget & assumptions

What must pass

At least 95% precision on a blinded, stratified assertion audit AND at least 10% lower error on a frozen anatomical task versus coarse-label and ancestry controls.

Why this probability

An operating corpus makes a useful resource plausible. Richer text may still add no information beyond ancestry or leak the target trait.

Stop or change direction

Stop scaling ingestion if extra description does not improve the frozen task or source fidelity fails. Accurate extraction alone is a narrower success.

Cost of the next check

Scientific work · 120 h × $100
$12.0k
Independent review · 40 h × $150
$6.0k
LLM allowance · 100M in / 20M out
$600
Compute, memory & storage allowance
$150
Other data costs allowance
$100
Contingency · 20%
$3.8k

Tools/data cash only: $1.0k including contingency, if labor is contributed. This excludes the value of human time. Actual historical spend: not metered here.

What a positive result could unlock

Broader source adjudication, licensing review and buyer validation: $50.0k–$150.0k additional planning allowance. No supplier quote; not included in the first test.

Potential customer and market

Evidence-linked anatomy data subscriptions and research integrations. Annual TAM scenario: $3.2M–$80.0M. Relevant buyers are assumed to be 10–50% of the 3,200-company US biotechnology R&D reference pool.

Annual price: $10.0k–$50.0k per organization. Neither qualification nor willingness to pay has been validated. Academic, non-US and clinical markets are excluded.

Commercial success is a separate event

25% conditional planning prior for winning three distinct paying organizations, each with a research pilot of at least $25,000, within 36 months of starting—assuming this gate passes and follow-on development and sales are funded. Joint chance under those assumptions: 11.25%.

This commercial prior has very low confidence and no customer-validation evidence. It is not the chance of a major biological breakthrough.

Related evidence ↗
04 / Partial signal

Sequence → knockout survival

Historical protein features reached AUROC 0.667 versus 0.525 for length on 1,639 later-measured mouse genes. This is partial ranking, not percent accuracy.

Next test · fully costed$10.4k$6.8k–$16.6k planning range · 2–4 weeks
Chance of passing this test40%20–60% judgment range · low confidence
Annual US TAM scenario$12.8M320 buyers × $40.0k/year
Test cost / annual TAM0.081%Scale comparison, not return on investment
Test, budget & assumptions

What must pass

At least +0.03 AUROC over the strongest applicable sequence baseline under both temporal and homology-family separation, with a positive paired 95% interval and acceptable calibration.

Why this probability

A measured temporal signal exists, but the earlier split did not exclude all homologous families. The sequence–essentiality relationship is established prior art.

Stop or change direction

Stop product claims if family separation removes the lift or errors remain too large for a defined research use.

Cost of the next check

Scientific work · 60 h × $100
$6.0k
Independent review · 16 h × $150
$2.4k
LLM allowance · 20M in / 4M out
$120
Compute, memory & storage allowance
$100
Other data costs allowance
$50
Contingency · 20%
$1.7k

Tools/data cash only: $324 including contingency, if labor is contributed. This excludes the value of human time. Actual historical spend: not metered here.

What a positive result could unlock

External-cohort replication and workflow utility study: $50.0k–$200.0k additional planning allowance. No supplier quote; not included in the first test.

Potential customer and market

Research-only gene prioritization for preclinical teams. Annual TAM scenario: $2.4M–$51.2M. Relevant buyers are assumed to be 5–20% of the 3,200-company US biotechnology R&D reference pool.

Annual price: $15.0k–$80.0k per organization. Neither qualification nor willingness to pay has been validated. Academic, non-US and clinical markets are excluded.

Commercial success is a separate event

15% conditional planning prior for winning three distinct paying organizations, each with a research pilot of at least $25,000, within 36 months of starting—assuming this gate passes and follow-on development and sales are funded. Joint chance under those assumptions: 6%.

This commercial prior has very low confidence and no customer-validation evidence. It is not the chance of a major biological breakthrough.

Related evidence ↗
05 / Strongest pilot signal

Integrated evidence → survival

Combined biological features reached AUROC 0.841–0.851 versus 0.744–0.771 for cell-fitness-only on fixed gene splits. These are a different cohort from the sequence pilot.

Next test · fully costed$16.8k$10.9k–$26.9k planning range · 3–5 weeks
Chance of passing this test45%25–65% judgment range · low confidence
Annual US TAM scenario$28.8M480 buyers × $60.0k/year
Test cost / annual TAM0.058%Scale comparison, not return on investment
Test, budget & assumptions

What must pass

At least +0.03 AUROC over the strongest single-source and published baseline on a genuinely external cohort, with improved calibration and a positive paired 95% interval.

Why this probability

The measured integration gain is encouraging. Transfer to independent measurements and superiority to existing tools remain untested.

Stop or change direction

Stop if date, source or gene-family leakage explains the lift, or the external cohort fails. An internal refit audit is not external biological replication.

Cost of the next check

Scientific work · 100 h × $100
$10.0k
Independent review · 24 h × $150
$3.6k
LLM allowance · 30M in / 6M out
$180
Compute, memory & storage allowance
$150
Other data costs allowance
$100
Contingency · 20%
$2.8k

Tools/data cash only: $516 including contingency, if labor is contributed. This excludes the value of human time. Actual historical spend: not metered here.

What a positive result could unlock

Prospective research utility and independent cohort validation: $75.0k–$250.0k additional planning allowance. No supplier quote; not included in the first test.

Potential customer and market

Evidence integration and prioritization for discovery teams. Annual TAM scenario: $4.0M–$115.2M. Relevant buyers are assumed to be 5–30% of the 3,200-company US biotechnology R&D reference pool.

Annual price: $25.0k–$120.0k per organization. Neither qualification nor willingness to pay has been validated. Academic, non-US and clinical markets are excluded.

Commercial success is a separate event

25% conditional planning prior for winning three distinct paying organizations, each with a research pilot of at least $25,000, within 36 months of starting—assuming this gate passes and follow-on development and sales are funded. Joint chance under those assumptions: 11.25%.

This commercial prior has very low confidence and no customer-validation evidence. It is not the chance of a major biological breakthrough.

Related evidence ↗
06 / Rework before scaling

Developmental failure timing

Accuracy was 54.2% versus 49.8% for length, but only 4 of 155 middle-stage cases were correctly classified. The cohort includes already-lethal mouse genes only.

Next test · fully costed$18.5k$12.0k–$29.5k planning range · 3–5 weeks
Chance of passing this test15%5–30% judgment range · low confidence
Annual US TAM scenario$5.6M160 buyers × $35.0k/year
Test cost / annual TAM0.330%Scale comparison, not return on investment
Test, budget & assumptions

What must pass

Middle-stage recall ≥30%, balanced accuracy ≥50%, and lower log loss than the strongest baseline on untouched domain-family groups, with a positive paired 95% interval for log-loss improvement.

Why this probability

The existing failure is substantial. Better context could help, but the available labels and class imbalance limit confidence.

Stop or change direction

Stop if better average accuracy continues to hide failure of the middle class. Do not extrapolate to arbitrary genes or human patients.

Cost of the next check

Scientific work · 100 h × $100
$10.0k
Independent review · 32 h × $150
$4.8k
LLM allowance · 40M in / 8M out
$240
Compute, memory & storage allowance
$250
Other data costs allowance
$100
Contingency · 20%
$3.1k

Tools/data cash only: $708 including contingency, if labor is contributed. This excludes the value of human time. Actual historical spend: not metered here.

What a positive result could unlock

Independent timing cohort and use-case validation: $75.0k–$250.0k additional planning allowance. No supplier quote; not included in the first test.

Potential customer and market

Developmental-stage prioritization for preclinical research. Annual TAM scenario: $1.2M–$24.0M. Relevant buyers are assumed to be 2.5–10% of the 3,200-company US biotechnology R&D reference pool.

Annual price: $15.0k–$75.0k per organization. Neither qualification nor willingness to pay has been validated. Academic, non-US and clinical markets are excluded.

Commercial success is a separate event

10% conditional planning prior for winning three distinct paying organizations, each with a research pilot of at least $25,000, within 36 months of starting—assuming this gate passes and follow-on development and sales are funded. Joint chance under those assumptions: 1.5%.

This commercial prior has very low confidence and no customer-validation evidence. It is not the chance of a major biological breakthrough.

Related evidence ↗
07 / Paused after negative result

Broad anatomy prediction

An initial 47-mammal pilot did not reliably beat both relatedness baselines. In a stronger 32-mammal test, adding gene profiles made error 0.43% worse than body mass plus relatedness.

Next test · fully costed$7.9k$5.1k–$12.7k planning range · 2–3 weeks
Chance of passing this test10%3–20% judgment range · low confidence
Annual US TAM scenario$4.0M160 buyers × $25.0k/year
Test cost / annual TAM0.198%Scale comparison, not return on investment
Test, budget & assumptions

What must pass

A changed representation reduces error by ≥10% beyond body mass plus relatedness on untouched families, with a positive paired 95% interval; original species are not a fresh confirmation set.

Why this probability

Two failed baseline comparisons lower the prior. Larger token budgets alone do not fix the information gap.

Stop or change direction

Do not spend this budget to rerun the same features. Reopen only with a materially different representation and a usable new holdout.

Cost of the next check

Scientific work · 40 h × $100
$4.0k
Independent review · 16 h × $150
$2.4k
LLM allowance · 10M in / 2M out
$60
Compute, memory & storage allowance
$80
Other data costs allowance
$50
Contingency · 20%
$1.3k

Tools/data cash only: $228 including contingency, if labor is contributed. This excludes the value of human time. Actual historical spend: not metered here.

What a positive result could unlock

Independent family replication, only after a positive gate: $50.0k–$150.0k additional planning allowance. No supplier quote; not included in the first test.

Potential customer and market

Comparative phenotype analysis for R&D groups. Annual TAM scenario: $800.0k–$16.0M. Relevant buyers are assumed to be 2.5–10% of the 3,200-company US biotechnology R&D reference pool.

Annual price: $10.0k–$50.0k per organization. Neither qualification nor willingness to pay has been validated. Academic, non-US and clinical markets are excluded.

Commercial success is a separate event

10% conditional planning prior for winning three distinct paying organizations, each with a research pilot of at least $25,000, within 36 months of starting—assuming this gate passes and follow-on development and sales are funded. Joint chance under those assumptions: 1%.

This commercial prior has very low confidence and no customer-validation evidence. It is not the chance of a major biological breakthrough.

Related evidence ↗
08 / Method check

Does LLM assistance add value?

LLMs helped write the existing workflows. No controlled experiment has measured their incremental scientific value or total billed token cost.

Next test · fully costed$10.0k$6.5k–$16.0k planning range · 2–3 weeks
Chance of passing this test30%15–50% judgment range · low confidence
Annual US TAM scenario$16.0M800 buyers × $20.0k/year
Test cost / annual TAM0.062%Scale comparison, not return on investment
Test, budget & assumptions

What must pass

At least 20% lower fully costed expense per correct sourced assertion versus a non-LLM workflow, at ≥95% precision on a blinded audit, with a positive paired 95% interval for savings.

Why this probability

Extraction and coding are plausible uses. Expert correction, source access and verification can consume the apparent savings.

Stop or change direction

Stop if savings disappear after expert correction, or if unsupported assertions increase. An LLM explanation is not an experimental observation.

Cost of the next check

Scientific work · 50 h × $100
$5.0k
Independent review · 20 h × $150
$3.0k
LLM allowance · 30M in / 6M out
$180
Compute, memory & storage allowance
$80
Other data costs allowance
$50
Contingency · 20%
$1.7k

Tools/data cash only: $372 including contingency, if labor is contributed. This excludes the value of human time. Actual historical spend: not metered here.

What a positive result could unlock

Independent workflow replication and customer pilot: $25.0k–$100.0k additional planning allowance. No supplier quote; not included in the first test.

Potential customer and market

Audited research-curation workflow for biotech teams. Annual TAM scenario: $1.6M–$64.0M. Relevant buyers are assumed to be 10–50% of the 3,200-company US biotechnology R&D reference pool.

Annual price: $5.0k–$40.0k per organization. Neither qualification nor willingness to pay has been validated. Academic, non-US and clinical markets are excluded.

Commercial success is a separate event

20% conditional planning prior for winning three distinct paying organizations, each with a research pilot of at least $25,000, within 36 months of starting—assuming this gate passes and follow-on development and sales are funded. Joint chance under those assumptions: 6%.

This commercial prior has very low confidence and no customer-validation evidence. It is not the chance of a major biological breakthrough.

Related evidence ↗
09 / LONG-TERM MISSION

A general genome-to-organism model

Reconstructing organism development from a genome has no sufficiently specified success criterion, time horizon or complete validation dataset in this program.

Decisive test costUnscoped
Probability of successNot estimable
Addressable marketUndefined

Assigning a precise percentage or a trillion-dollar market would manufacture confidence. The bounded tests above are the current way to buy evidence toward this mission. A positive result on one does not establish a universal DNA language.

ASSUMPTION LAB

Make the tradeoff explicit.

Choose a path, then adjust its next-test budget and market assumptions. Edits affect this calculator only and reset when you reload.

Cost to check
Opportunity and likelihood

Commercial milestone: 3 organizations, each paying ≥$25k for a research pilot, by month 36. Conditional on funded follow-on development and sales. The year-3 revenue scenario is separate; it is not probability-weighted.

Annual market capacity and a one-time test cost have different time bases. Their ratio is a scale comparison. It omits future product development, sales, delivery expenses, taxes and financing; it is not ROI or expected value.

READ THE ASSUMPTIONS

Evidence before arithmetic.

1 / Fully costed research

Test cost = (scientific hours × rate + review hours × rate + input/output tokens × token rates + compute/storage + other data costs) × (1 + contingency). Scope is public-data curation, modeling, controls and review. No new wet-lab experiments, genome-wide foundation-model training or clinical validation are included.

Default $100/hour scientific labor is a rounded planning assumption: the BLS R&D-industry medical-scientist median of $125,700/year ÷ 2,080 hours × an assumed 1.65 burden factor ≈ $100/hour. Reviewer $150/hour, work hours, 20% contingency and 0.65–1.6× budget ranges are assumptions, not quotes. Contributed labor is still an economic cost. BLS, May 2025 wages ↗

Token rates of $3/M input and $15/M output are editable budget assumptions, not a named model quote or actual historical bill. Reasoning/retry tokens must be included when metered. We have not reconstructed past subscription or token spend.

Compute allowances cover small runs, not training a frontier genome model. As a reference, Modal lists A100 80 GB at $0.000694/second (about $2.50/GPU-hour), with CPU, memory and storage billed separately. Each path’s allowance also includes those resources; it is not a reserved purchase. Modal resource pricing, checked September 12, 2026 ↗

2 / A market you can challenge

Annual US TAM scenario = qualified buyer organizations × annual price per organization. The starting reference pool is 3,200 US companies performing biotechnology R&D in the 2023 NSF/Census BERD survey. The table flags that count as substantially imputed/reweighted and excludes companies with fewer than 10 domestic employees. It is not a list of customers.

Which fraction needs each proposed product, and what it would pay, are unvalidated assumptions. The survey reports about $136.1B in biotechnology R&D spending; that is industry context, not DNALYZER’s TAM. NSF/Census BERD 2023, table 19 ↗

Low/base/high TAM combines low/base/high buyer and price assumptions. Ranges are sensitivity scenarios, not confidence intervals. These products compete for overlapping research budgets; do not sum their TAMs. Academic and international buyers, therapeutic sales, royalties and clinical diagnostics are outside this initial model.

Before treating any market as validated: enumerate qualified organizations, interview budget owners, test price and obtain paid pilots. No customer interviews or signed demand are claimed here.

3 / Define success before assigning odds

Each research probability means passing all criteria stated for its next test, within the indicated 2–8 week execution window after data access and staffing. It is conditional on that budget, usable data, legal access and expert review. Freeze the detailed metric, data split, sample-size justification and analysis before observing new results.

Ranges express judgment uncertainty; they are not statistical confidence intervals. No calibration dataset or elicited expert panel exists for these initial estimates. A published precedent supports feasibility, not a numerical success rate. Positive pilots raise the prior; weak controls, sparse labels and failed replications lower it.

Commercial joint probability = P(test passes) × P(3 paid pilots by month 36 | test passes and follow-on funded). This is a conditional chain, not an independence assumption, and it is not a probability of scientific breakthrough, profitability or company survival.

4 / Update instead of defending a number

Record each forecast, its assumptions, the frozen outcome definition and the eventual result. Update after data-readiness checks, independent evaluation and buyer interviews. Preserve the old estimate with the reason for the revision. Until enough comparable outcomes exist, avoid calling the model calibrated.

A failed hypothesis can still be a successful use of a small budget if it prevents a much larger uninformative program. This model therefore shows stop rules and follow-on costs rather than ranking solely by market size.

Follow-on allowances describe research validation only. Total cost to a commercial product remains unscoped and may materially exceed those amounts. Full proprietary protocols, candidate rankings, code and raw datasets remain private.

Download the dated assumptions

Earlier pilots in this portfolio

The sequence, integrated-evidence and timing rows summarize DNALYZER’s September 11, 2026 computational pilots. Their local refit and metric audits check implementation; they do not establish independent biological replication or novelty. AUROC values from different cohorts are not head-to-head comparisons.

Source measurements came from the IMPC release archive, Cacheiro et al. (2020) and the published embryonic timing dataset. These sources support the underlying measurements; the numerical model results are DNALYZER’s analyses, not results attributed to those authors.

These are early public summaries, not peer-reviewed papers or complete reproduction packages. No new biological mechanism or universal DNA decoder is claimed.