Repository navigation

Sync local main: README revision, DE-59 renumber, gitignore cleanup - #111

Merged

jam-sudo merged 233 commits into

mainfrom

docs/readme-revision-2026-10-01

Oct 2, 2026

Merged

Sync local main: README revision, DE-59 renumber, gitignore cleanup#111
jam-sudo merged 233 commits into
mainfrom
docs/readme-revision-2026-10-01

Commits on Sep 23, 2026
- fix: harden public prediction and deployment contracts
  jam-sudo
  committed
  e8dcd12
- fix: validate finite holdout data and smoke-test API image
  jam-sudo
  committed
  a266e7a
- build: add reproducible holdout evaluator container
  jam-sudo
  committed
  baeb763
- fix: lock external holdout scoring and corpus integrity
  jam-sudo
  committed
  ab13a08
- fix: bind holdout acquisition plan to verified cohort files
  jam-sudo
  committed
  6faf5db
- docs: quantify official regulatory inventory feasibility
  jam-sudo
  committed
  2d61e28
- fix: enforce preregistered dates for clinical Cmax sources
  jam-sudo
  committed
  6b4a31b
- docs: quantify NCATS FRDB holdout candidate attrition
  jam-sudo
  committed
  f8074e3
- docs: record EPA Cmax extract screening limits
  jam-sudo
  committed
  4c34255
- fix: freeze holdout verifier provenance
  jam-sudo
  committed
  b7a401f
- fix: pin holdout resources to frozen checkout
  jam-sudo
  committed
  0534dc0
- docs: assess Health Canada Cmax source yield
  jam-sudo
  committed
  f95818f
- test: isolate N50 freeze fixture and enforce recovered statin gate
  jam-sudo
  committed
  b6461b8
- docs: screen ClinicalTrials.gov Cmax source feasibility
  jam-sudo
  committed
  f71cd60
- docs: quantify structural attrition in trial-registry leads
  jam-sudo
  committed
  ce13329
- docs: assess PMDA review-report candidate pool
  jam-sudo
  committed
  cdf72b4
- fix: show actual prediction evidence in console
  jam-sudo
  committed
  780e31c
- fix: enforce fallback holdout superiority margin
  jam-sudo
  committed
  1cd4c2f
- fix: align regimen dose boundaries and mass balance
  jam-sudo
  committed
  5f86e72
- fix: preserve regimen AUC across dose jumps
  jam-sudo
  committed
  3b6d8fd
- fix: enforce regimen horizons after infusion ends
  jam-sudo
  committed
  bc9ed5f
- docs: rule out local DrugBank text as holdout source
  jam-sudo
  committed
  5a2ada9
- docs: describe VDss track as volume proxy
  jam-sudo
  committed
  98dbe61
- docs: distinguish archived preprint from current benchmark
  jam-sudo
  committed
  618106b
- docs: preregister self-run external-data pilot
  jam-sudo
  committed
  8943113
- docs: align pilot screen with FRDB structure availability
  jam-sudo
  committed
  34ba64a
- validation: freeze label-free P0 candidates
  jam-sudo
  committed
  5392bdd
- validation: freeze P0 prediction runner and environment
  jam-sudo
  committed
  667c717
- validation: seal P0 predictions before outcome extraction
  jam-sudo
  committed
  84856cf
- validation: record provisional first-ten source review
  jam-sudo
  committed
  be863ce
- validation: correct healthy-elderly source adjudication
  jam-sudo
  committed
  f6bf5b4
- validation: cross-check second P0 source batch
  jam-sudo
  committed
  ba5c1c6
- validation: adjudicate third P0 source batch
  jam-sudo
  committed
  c5ff8e3
- Record blinded P0 source review for candidates 31-40
  jam-sudo
  committed
  cd425ea
- validation: adjudicate fourth P0 source batch
  jam-sudo
  committed
  754df2b
- validation: adjudicate fifth P0 source batch
  jam-sudo
  committed
  edda094
- Record P0 source review for candidates 61-70
  jam-sudo
  committed
  7728c34
- Record P0 source review for candidates 71-80
  jam-sudo
  committed
  4c5d7b3
- Record blinded P0 source review for candidates 81-90
  jam-sudo
  committed
  38020b6
- Record blinded P0 source review for candidates 91-100
  jam-sudo
  committed
  9ebc7a2
- Record blinded P0 source review for candidates 101-110
  jam-sudo
  committed
  7ac82f4
- Record blinded P0 source review for candidates 111-120
  jam-sudo
  committed
  3c4d7e9
- Record blinded P0 source review for candidates 121-130
  jam-sudo
  committed
  dcec0cc
- Record blinded P0 source review for candidates 131-140
  jam-sudo
  committed
  9cc274b
- Record blinded P0 source review for candidates 141-150
  jam-sudo
  committed
  5ed4a75
- Record blinded P0 source review for candidates 151-160
  jam-sudo
  committed
  0a10649
- Record blinded P0 source review for candidates 161-170
  jam-sudo
  committed
  8a6bbeb
- Record blinded P0 source review for candidates 171-180
  jam-sudo
  committed
  ad89990
- Record blinded P0 source review for candidates 181-186
  jam-sudo
  committed
  02205aa
- Freeze P0 source adjudication rules before prediction unsealing
  jam-sudo
  committed
  df6c37f
- Freeze P0 source adjudication before prediction unsealing
  jam-sudo
  committed
  d9c4e4c
- Score frozen P0 cohort and document diagnostic result
  jam-sudo
  committed
  1093633
- Record P0 track diagnostics and post-unseal source concern
  jam-sudo
  committed
  6f41640
- Disclose diagnostic P0 result in project validation claims
  jam-sudo
  committed
  90a1154
- Require verified parent dose basis in V1 holdout contract
  jam-sudo
  committed
  ca84507
- Verify post-dose fasting duration in V1 holdout
  jam-sudo
  committed
  61aa49b
- Connect P0 bioavailability diagnosis to prior failed levers
  jam-sudo
  committed
  3d74d35
- Report P0 source-statistic sensitivity analyses
  jam-sudo
  committed
  f8a9a9b
- Bind source Cmax statistic in V1 holdout labels
  jam-sudo
  committed
  861a88c
- Avoid eager NumPy trapz fallback evaluation
  jam-sudo
  committed
  94fd4c3
- Trace P0 ipragliflozin exposure error to gut uptake
  jam-sudo
  committed
  db83d40
- Classify phenolic amines by strong ionizable groups
  jam-sudo
  committed
  ffb62fb
- Document ipragliflozin formulation sensitivity diagnostic
  jam-sudo
  committed
  5e095a4
- Commit blind holdout outcomes before prediction
  jam-sudo
  committed
  4b9bfe0
- Seal predictions and isolate final-test holdout cohort
  jam-sudo
  committed
  64f8b77
- Audit declared holdout synonyms and drug relations
  jam-sudo
  committed
  9156f24
- Align holdout power claim and guard curated relation fields
  jam-sudo
  committed
  a138179
- Resolve early IV exposure for bioavailability estimates
  jam-sudo
  committed
  1893287
- Use converged exposure for measured bioavailability
  jam-sudo
  committed
  ded0a0b
- Fail closed in N50 exclusion audits
  jam-sudo
  committed
  19053b7
- Normalize salt forms in holdout exclusion paths
  jam-sudo
  committed
  885c52a
- Require verifiable training sources for external holdout
  jam-sudo
  committed
  7840abb
- Quantify blinded holdout identity source overlap
  jam-sudo
  committed
  860ac08
- Pin recovered Omega Cmax source for exclusion audits
  jam-sudo
  committed
  966a680
- Pin co-committed CLF and VDF training source
  jam-sudo
  committed
  f0c4bd4
- Treat DrugBank fup overlap as prospective exclusion
  jam-sudo
  committed
  cac1ba6
- Block DrugBank-trained fup artifact from public holdout
  jam-sudo
  committed
  3ff9cd3
- Retrain public fup model and refresh development evidence
  jam-sudo
  committed
  7f9790c
- Pin public Peff training and refresh current predictions
  jam-sudo
  committed
  f60da3c
Commits on Sep 24, 2026
- Pin public Cmax training and refresh current evidence
  jam-sudo
  committed
  6968ef7
- Pin public hepatocyte CLint training and refresh evidence
  jam-sudo
  committed
  c8f0b09
- Pin public VDss training and bind current evidence
  jam-sudo
  committed
  803e8a5
- Bind external score to frozen audit and cohort target
  jam-sudo
  committed
  5ab7c75
- Reject analyte mismatch and verify frozen runtime settings
  jam-sudo
  committed
  ef8fe7f
- Report paired Meta versus ML uncertainty on development cohort
  jam-sudo
  committed
  324b81e
- Fail closed on Peff model errors and nonfinite ADME values
  jam-sudo
  committed
  30332e3
- Preflight residual calibration and fail strict VDss errors
  jam-sudo
  committed
  9c46e54
- Document PK-DB external holdout feasibility screen
  jam-sudo
  committed
  2c2f52d
- Limit development residual interval to calibrated prediction path
  jam-sudo
  committed
  9e02af0
- Verify residual calibration provenance before displaying interval
  jam-sudo
  committed
  3786fe0
- Align research claims with current validation evidence
  jam-sudo
  committed
  aa0a9a4
- Reject superiority gate for undersized external cohorts
  jam-sudo
  committed
  ed0380a
- Quantify FDA oral NME name-screen attrition
  jam-sudo
  committed
  c2c7fb1
- Correct pravastatin parent identity and refresh development evidence
  jam-sudo
  committed
  1da8bcd
- Adjudicate development stereochemistry and pin UGT routing
  jam-sudo
  committed
  827d276
- Bind experimental docking caches to exact molecular structure
  jam-sudo
  committed
  0e05bd7
- Exclude modeler-seen FRDB leads from blinded holdout
  jam-sudo
  committed
  26634af
- Remove pravastatin leakage from production CLint training
  jam-sudo
  committed
  8dd09d9
- Guard all production fitted snapshots against development overlap
  jam-sudo
  committed
  6933949
- Qualify P0 diagnostic for possible VDss training overlap
  jam-sudo
  committed
  e32250d
- Require the pinned public fup model at runtime
  jam-sudo
  committed
  a45265f
- Verify fitted artifact hashes before external holdout
  jam-sudo
  committed
  89231e9
- Screen two additional public PK acquisition sources
  jam-sudo
  committed
  7b474d5
- Screen e-Drug3D for untouched oral Cmax leads
  jam-sudo
  committed
  d7e2508
- Make direct DrugBank lookups respect execution profile
  jam-sudo
  committed
  a56c8a4
- Fail closed on licensed logP feature drift
  jam-sudo
  committed
  a158416
- Adjudicate amphetamine reference identity and dose
  jam-sudo
  committed
  316f075
- Adjudicate morphine development reference and refresh evidence
  jam-sudo
  committed
  d2dcebd
- Adjudicate Baekdal digoxin and quarantine warfarin reference
  jam-sudo
  committed
  b56f427
- Adjudicate remaining PMID-linked training references
  jam-sudo
  committed
  c5e2a61
- Quarantine mismatched PMID-linked Cmax references
  jam-sudo
  committed
  0e01e5d
- Complete PMID-linked training reference adjudication
  jam-sudo
  committed
  3be35c6
- Audit estimated development Cmax labels and refresh benchmark
  jam-sudo
  committed
  27dbc69
- Correct prodrug development labels and preserve holdout leakage guard
  jam-sudo
  committed
  320cc29
- Quarantine three more prodrug metabolite development labels
  jam-sudo
  committed
  1191582
- Correct mismatched development Cmax dose arms
  jam-sudo
  committed
  0cc7468
- Quarantine steady-state labels from single-dose development benchmark
  jam-sudo
  committed
  90a0147
- Audit four additional development PK reference arms
  jam-sudo
  committed
  6839583
- Replace three generic PK arms with exact observed sources
  jam-sudo
  committed
  5c2b4db
- Correct dasatinib development Cmax to single-dose arm
  jam-sudo
  committed
  657c300
- Audit development calibration reference arms
  jam-sudo
  committed
  00d2390
- Correct additional calibration labels and bind interval sources
  jam-sudo
  committed
  786cca6
- Screen development references for analyte and regimen mismatches
  jam-sudo
  committed
  613fbf5
- Audit training dose and specimen provenance
  jam-sudo
  committed
  f3c4da1
- Correct five development PK arms and source inventory
  jam-sudo
  committed
  2e004b8
- Re-source six development PK training arms
  jam-sudo
  committed
  60cf5e9
- Correct three more training PK source mismatches
  jam-sudo
  committed
  0074cea
- Correct four more development PK reference arms
  jam-sudo
  committed
  ea04f38
- Re-source ibuprofen and metoclopramide training arms
  jam-sudo
  committed
  87fe5b2
- Re-source nifedipine and pitavastatin development arms
  jam-sudo
  committed
  fd603d5
- Replace steady-state development proxies with observed single-dose arms
  jam-sudo
  committed
  e999d6d
- Clarify four development label arms and clomipramine dose basis
  jam-sudo
  committed
  e430461
- Reconcile current development metrics in experiment log
  jam-sudo
  committed
  192fed8
- Audit additional development PK source arms
  jam-sudo
  committed
  a3f13a7
- Correct oral PK arms against regulatory labels
  jam-sudo
  committed
  fd84567
- Quarantine four mismatched development PK arms
  jam-sudo
  committed
  53f29ad
- Correct generic-label development PK arms
  jam-sudo
  committed
  a5f6a5b
- Audit remaining generic-label development PK arms
  jam-sudo
  committed
  e2c805d
- Verify OSP observed analyte and single-dose provenance
  jam-sudo
  committed
  882251d
- Correct acamprosate and phenytoin development references
  jam-sudo
  committed
  908a484
- Remove synthetic clinical reference curves
  jam-sudo
  committed
  1e9e3dd
- Bind external holdout names to verified shortlist
  jam-sudo
  committed
  7c61002
- Remove stale headline pins from probe isolation tests
  jam-sudo
  committed
  77ef337
- Correct scored parent doses and quinine source arm
  jam-sudo
  committed
  be3b44c
- Replace ambiguous codeine reference with FDA fasted arm
  jam-sudo
  committed
  ba0e8c7
- Correct implicit salt doses and mismatched PK arms
  jam-sudo
  committed
  2d390d6
- Adjudicate remaining salt strengths and PK source arms
  jam-sudo
  committed
  7f2f576
- Correct ulipristal acetate reference identity and preserve exclusion lineage
  jam-sudo
  committed
  369bc68
- Use original dapagliflozin Cmax table for development reference
  jam-sudo
  committed
  558ce69
- Audit OSP silver PK references and preserve quarantines
  jam-sudo
  committed
  e99d3a9
- Reconcile three silver PK arms with original sources
  jam-sudo
  committed
  e978ff0
- Quarantine penicillamine midpoint and verify three PK arms
  jam-sudo
  committed
  8a69a25
- Quarantine acamprosate peak and correct three PK source arms
  jam-sudo
  committed
  b463bb6
- Quarantine repeat-dose alvimopan and align donepezil parent dose
  jam-sudo
  committed
  b32be66
- Correct ketorolac development arm and trace brincidofovir source
  jam-sudo
  committed
  2102026
- Correct lamivudine single-dose reference and mercaptopurine provenance
  jam-sudo
  committed
  f989d16
- Quarantine repeat-dose progesterone and retire stale FDA builder
  jam-sudo
  committed
  f40704e
- Replace indomethacin benchmark arm with primary regulatory data
  jam-sudo
  committed
  cbe72b8
- Use original fasted ketoconazole tablet Cmax arm
  jam-sudo
  committed
  59774a5
- Guard public benchmark figures against cache drift
  jam-sudo
  committed
  dfd82d0
- Document metadata-only ClinicalTrials source screen
  jam-sudo
  committed
  b48def6
- Triage protocol-only trial leads without opening outcomes
  jam-sudo
  committed
  0418d77
- Align F diagnostic with converged production calculation
  jam-sudo
  committed
  50dd25e
- Correct absolute bioavailability diagnostic provenance
  jam-sudo
  committed
  ac8123f
- Correct overstated Cmax ceiling diagnosis and public audit notes
  jam-sudo
  committed
  4f114c8
- Restore CI lint for reference regression checks
  jam-sudo
  committed
  95d215d
- Include runtime schema validator in API image
  jam-sudo
  committed
  86afaa5
- Align public limitations with source-verified evidence
  jam-sudo
  committed
  edb246b
- Screen PKRxiv as external holdout source
  jam-sudo
  committed
  0f8d601
- Clarify empirical absorption units and latent transporter scope
  jam-sudo
  committed
  3d97105
- Recompute frozen holdout audit and index keyed drug names
  jam-sudo
  committed
  7aed581
- Exclude modeler-seen ClinicalTrials leads from final holdout
  jam-sudo
  committed
  84c0d3c
- Freeze AI-curated blind P1 diagnostic predictions before label reveal
  jam-sudo
  committed
  40d222e
- Report source-checked blind P1 diagnostic and limitations
  jam-sudo
  committed
  79ba057
- Index registry aliases and record blind pilot interval gaps
  jam-sudo
  committed
  f480b84
- Document P1 structural-neighbor diagnostic
  jam-sudo
  committed
  737d344
- Trace blind pilot exposure errors and correct Cmax model description
  jam-sudo
  committed
  ddbfdf2
- Clarify rule-based applicability and direct Cmax tail errors
  jam-sudo
  committed
  9541ca7
- Record exhausted FDA P2 acquisition window
  jam-sudo
  committed
  ea21bac
- Clarify P2 source exclusions and release follow-up worker
  jam-sudo
  committed
  1ec52d7
- Exclude P2 modeler-seen parents from external holdout
  jam-sudo
  committed
  5bc7259
- Quantify older FDA identity pool without Cmax outcomes
  jam-sudo
  committed
  9298c77
- Record older FDA source availability pilot
  jam-sudo
  committed
  f6fb5e9
- Correct felbamate Cmax source unit and rebuild public evidence
  jam-sudo
  committed
  8a95a5c
Commits on Sep 25, 2026
- Refresh accuracy diagnosis and record chiral ablation
  jam-sudo
  committed
  df8deb6
- Correct older FDA holdout identity screen and salt-name gate
  jam-sudo
  committed
  8a2cd82
- Update PMDA identity feasibility after salt normalization
  jam-sudo
  committed
  6c17c59
- Correct P1 diagnostic for whole-blood Cmax arm
  jam-sudo
  committed
  ad9f240
- Reject uncertain source labels during holdout integration
  jam-sudo
  committed
  8c19c35
- Require exact DrugBank names in PK integration
  jam-sudo
  committed
  94f6993
- Quarantine whole-blood indapamide Cmax training labels
  jam-sudo
  committed
  bcfd986
- Quarantine immunosuppressant blood Cmax labels
  jam-sudo
  committed
  cc8eb99
- Quarantine administered-analyte mismatches in Cmax training
  jam-sudo
  committed
  f66dd68
- Quarantine pimecrolimus blood Cmax training label
  jam-sudo
  committed
  78dce35
- Quarantine voclosporin whole-blood Cmax training label
  jam-sudo
  committed
  eb1d257
- Report scaffold CV for direct Cmax model
  jam-sudo
  committed
  e538574
- Pin both Omega Cmax source workbooks
  jam-sudo
  committed
  898ea66
- Preserve one-time holdout prediction output
  jam-sudo
  committed
  3bf451f
- Document phenylbutyrate salt-dose adjudication
  jam-sudo
  committed
  ebc9edf
- Preserve one-time external holdout score output
  jam-sudo
  committed
  d0b8206
- Quarantine dolasetron metabolite-derived training labels
  jam-sudo
  committed
  16414e9
- Reconcile holdout protocol with screened Cmax artifact
  jam-sudo
  committed
  216ed69
- Screen CLF and VdF training and correct two salt-dose labels
  jam-sudo
  committed
  788aab7
- Correct four additional salt-basis training doses
  jam-sudo
  committed
  c360138
- Quarantine methenamine concentration-basis labels
  jam-sudo
  committed
  b4bc10a
- fix(data): normalize acoramidis hydrochloride dose
  jam-sudo
  committed
  55a116d
- fix(data): align givinostat paired mass basis
  jam-sudo
  committed
  07ba8bd
- docs(validation): adjudicate supplemental PK cohorts
  jam-sudo
  committed
  f6e1b7c
- fix(validation): bind holdout scoring to frozen release
  jam-sudo
  committed
  60a0757
- docs(validation): record amd64 holdout smoke
  jam-sudo
  committed
  e390005
- docs(validation): clarify ASA and ezetimibe source checks
  jam-sudo
  committed
  7703923
- docs(validation): verify trospium source arm with FDA review
  jam-sudo
  committed
  e68dfde
- docs(validation): verify metformin 1981 source arm
  jam-sudo
  committed
  019e0ef
- docs(validation): confirm metformin 1979 dose basis in FDA review
  jam-sudo
  committed
  f90119d
- docs(validation): verify two supplemental Cmax arms
  jam-sudo
  committed
  59636a7
- fix(data): normalize almitrine bismesylate dose
  jam-sudo
  committed
  1dfdc06
- docs(validation): reconcile current Cmax metrics and source counts
  jam-sudo
  committed
  cc339ba
- fix(data): normalize benazepril hydrochloride dose
  jam-sudo
  committed
  9b8f0b7
- fix(data): normalize orphenadrine hydrochloride dose
  jam-sudo
  committed
  10b6258
- fix(data): normalize bisoprolol hemifumarate doses
  jam-sudo
  committed
  3abb68a
- docs(validation): retract unsupported accuracy ceilings
  jam-sudo
  committed
  5b4c9c3
- fix(data): restore reference stereo and residual interval provenance
  jam-sudo
  committed
  f19d463
- fix(console): correct preset molecular identities and deployed data
  jam-sudo
  committed
  27d7414
- Downgrade levocetirizine reference confidence after source cross-check
  jam-sudo
  committed
  1040ae4
- Restore residual interval provenance after reference update
  jam-sudo
  committed
  14f3a52
- Screen new public PK benchmark for external holdout feasibility
  jam-sudo
  committed
  f91c5c3
Commits on Sep 26, 2026
- Record direct Cmax source metadata ablation no-go
  jam-sudo
  committed
  de2fd79
Commits on Oct 1, 2026

Commits

Commits on Sep 23, 2026

Commits on Sep 24, 2026