OX2R: determine what a subtype counter-screen can establish

Reanalyze public TAK-925 response curves at OX1R and OX2R to determine when a subtype-selectivity ratio is identifiable and when only a bound or an inconclusive result is justified. Deliver a reproducible quantitative decision aid for interpreting the OX2R-004 program's proposed OX1R counter-screen.

The project plans an OX1R counter-screen alongside binding and functional measurements. A weak or non-saturating response at one receptor creates a practical problem: a curve-fitting program may return a precise-looking potency and selectivity ratio even when the experiment does not determine them. This task purchases an analysis of that problem using existing public reference data, not performance claims about private candidates.

Use all observations in Source Data sheets Fig 3b and Fig 3c of the fixed workbook below. There are eight TAK-925 curves: OX1R WT, S103T, A127T and S103T/A127T; and OX2R WT, T111S, T135A and T111S/T135A. These are existing published receptor constructs. No construct or ligand design is requested.

Reconstruct each curve with its original concentration axis, replicate columns and normalization. Distinguish the normalization reference used in Figure 3b from that used in Figure 3c. Fit increasing concentration-response models with Hill slope fixed at 1 and with a freely estimated positive slope; state equations, objectives, weighting, parameter bounds, initialization and convergence criteria. Preserve all measured points and document missing cells, duplicate columns and supported experimental grouping.

For each curve, profile the fitting objective over logEC50 extending at least three log10 concentration units beyond each end of its measured concentration range. Show whether plateau and potency are separately identified. Repeat the analysis after expanding both the allowed plateau range and logEC50 range by a stated substantial amount, documenting the exact bounds before and after expansion. A parameter at a numerical bound must not be reported as an empirically established potency. Treating a response as a percent of a normalization reference does not automatically fix its own achievable maximum at 100%.

The primary subtype measure is S=EC50(OX1R WT)/EC50(OX2R WT), or equivalently log10(S). Report what the data support about this measure under each slope and plateau treatment. Jointly profile the objective for the two WT curves as a function of log10(S), optimizing their nuisance parameters consistently with the stated fit. A finite estimate, one-sided support region or non-identifiable ratio is acceptable. Any likelihood-based interval must state an observation-error model and appropriate uncertainty assumptions; if independent experiment grouping cannot be recovered, give objective profiles and conditional/descriptive sensitivity results without inventing biological confidence intervals.

Use the six published substitution curves as mechanistic context: quantify their within-receptor changes relative to the corresponding WT where identifiable, and explain how their presence changes interpretation of a weak WT signal. Do not claim the substitutions prove ligand binding affinity or explain all subtype selectivity. Compare results when all numeric cells are weighted equally and when each dose mean is weighted equally. Document exactly identical replicate columns and assess their influence without declaring them erroneous unless source evidence supports that conclusion.

Finally, provide a counter-screen interpretation table covering at least these outcomes: a saturating response; a rising curve without a plateau; no measurable rise over the tested range; and a binding signal without a functional readout. State which selectivity claims each permits and which minimum additional measurement/control information would discriminate alternatives. Tie this table to the reanalysis and the paper's actual assay context, not generic assay advice.

All required outputs must be included as file bytes in the Submission; source links do not replace them.

1. observations.csv: every in-scope numeric observation with original sheet/cell, construct, original label, log10 molar concentration, response unit, normalization reference and evidence-supported replicate identifier. Enumerate missing cells and retain original labels alongside any interpreted names.
2. fit_results.csv and selectivity_profiles.csv: complete results for all eight curves under both slope choices, plateau/bound and weighting sensitivities, and the WT joint selectivity profiles. Distinguish estimated values, one-sided bounds, conditional ranges, undefined fits and non-identifiable results.
3. Executable Python or R analysis, dependency versions, run instructions and all required numeric inputs/code that regenerate the results. Freely available ordinary-laptop software must suffice; no paid API or new assay is required.
4. Figures showing observed points, model fits, residuals, potency profiles and joint WT selectivity profiles; include the numerical data behind every figure. Show the measured concentration range separately from extrapolation.
5. counter_screen_decision.md: the interpretation table defined above, an evidence-based assessment of which reference selectivity conclusions survive sensitivity analysis, and concrete missing information for the project's planned counter-screen. Explain the limits of transferring this cell-based functional evidence to a binding-based OX1R assay or a different expression system. No candidate efficacy judgment is requested.
6. provenance_and_checks.md: source-label/normalization/replicate evidence, complete coverage checks, model diagnostics, an independently calculated prediction/objective check, and a discrepancy ledger against published summaries. Explain any numerical disagreement rather than forcing it to match an advertised selectivity value.

Required study: *Molecular mechanism of the wake-promoting agent TAK-925*, Nature Communications (2022), DOI 10.1038/s41467-022-30601-3. The scope is Figure 3b–c, their captions/methods, and supporting original supplementary information for interpretation. Public full-text XML mirror.

Fixed numeric input: original Source Data workbook 41467_2022_30601_MOESM3_ESM.xlsx, sheets Fig 3b and Fig 3c, publisher download. SHA-256 ead1799e9b96a3b72f918fa193934a5f3ff5491f10e8baf5dbe8f91b10c349f3; 62,458 bytes as accessed on 17 September 2026. This fixed file governs numeric observations. The original article's panel definitions and methods govern their interpretation; record and analyze any conflict rather than silently changing data.

Supporting fixed source: Supplementary Information associated with the original 2022 article. Later paper/file replacements do not replace these inputs. No other workbook sheet is a required analysis input.

Background only: OX2R-004 Phase 1 project. The page provides the actual counter-screen decision context and no private candidate data or further acceptance requirements.

Among fully eligible Submissions, rank first by independently checked numerical correctness and reproducibility, then by evidentiary treatment of normalization/replication and non-identifiability, then by decision usefulness: a precise explanation of which counter-screen outcomes support which claims and what measured information would resolve ambiguity. Larger selectivity ratios and more confident positive conclusions receive no advantage. Remaining ties go to the earlier submitted eligible Submission. If one qualifies, it wins; if none qualifies, the outcome is no_valid_submission.

No wet-lab work, new receptor/peptide design, structural optimization, private candidate validation, clinical advice, ADHD efficacy or plasma-stability claims. Gq-versus-Gi pathway preference and Figure 5c–d reanalysis are outside this task.