Reject a geometric mean over a zero ratio

Review ask

As a clinical trial statistician or programmer, answer two short questions in Your review: the task and the inputs. Agreeing with both approves this benchmark; disagreeing with either needs a comment. Propose a fix with the Edit button on any source file, or add it to your comments.

We suggest spending no more than 5 minutes on a review. The other tabs beside Your review are background for assessment, not under review.

Summary

EditUnder reviewWhat this benchmark means

README.md

Public entries: Python and R. Passing coverage requires an installed inventory.

Goal: carry on each subject-level record MRGM, the geometric mean of the subject's metabolite-to-parent ratios of maximum concentration across periods.

Input: demographics with one record per subject, and the pharmacokinetic parameters with one metabolite ratio per subject and period, one of them zero because the metabolite concentration was below the limit of quantification.

Variables:

MRGM would be the exponential of the mean of the natural logarithms of the subject's ratios. A ratio of zero has no logarithm, so no geometric mean exists for that subject, and the run fails while the values are computed and no artifact is accepted.

How to fix

Decide what a ratio of zero means for the analysis. The usual geometric mean is taken over the ratios above zero, which leaves the zero ratio out:

- name: MRGM
  type: float
  derivation:
    aggregate:
      key: [STUDYID, USUBJID]
      filter: "PP.PPTESTCD = 'MRCMAX' AND PP.PPSTRESN > 0"
      expr: "GEOMEAN(PP.PPSTRESN)"

If the study instead substitutes a value related to the limit of quantification, derive that stated substitute first and take the geometric mean of it. Do not replace zero with an unstated constant.

Expected failure

Editthe run is rejected; no artifact is accepted

expected/error.yaml

phase: derivation
condition: ln_of_nonpositive
rule: FORMULA-15

expected/error.yaml

phase: derivation
condition: ln_of_nonpositive
spec_paths:
  - columns.MRGM.derivation.aggregate.expr
rule: FORMULA-15
context:
  expr: "GEOMEAN(PP.PPSTRESN)"
  keys:
    - STUDYID: YAMAA-01
      USUBJID: YAMAA-01-S02

YAML specification

1 spec file

spec.yaml

Edit40 lines

1yamaa: "1.0"2type: spec3model: ADaM4domain: ADSL5keys: [STUDYID, USUBJID]67input:8  DM: input/dm.csv9  PP:10    path: input/pp.csv11    types:12      PPSEQ: int13      PPSTRESN: float14base: DM1516output:17  path: adsl.csv18  columns: [STUDYID, USUBJID, MRGM]1920columns:21  - name: STUDYID22    type: str23    label: Study Identifier24    derivation: DM.STUDYID2526  - name: USUBJID27    type: str28    label: Unique Subject Identifier29    derivation: DM.USUBJID3031  # The geometric mean takes the logarithm of each ratio. A ratio of zero has32  # none, so the run fails while values are computed.33  - name: MRGM34    type: float35    label: Geometric Mean Metabolite Ratio Cmax36    derivation:37      aggregate:38        key: [STUDYID, USUBJID]39        filter: "PP.PPTESTCD = 'MRCMAX'"40        expr: "GEOMEAN(PP.PPSTRESN)"

Inputs

Under review2 source files

input/dm.csv

Edit2 rows

input/dm.csv
STUDYID	USUBJID
YAMAA-01
YAMAA-01

input/pp.csv

Edit4 rows

input/pp.csv
STUDYID	DOMAIN	PPSEQ	PPTESTCD	PPTEST	PPSTRESN	PPSTRESU	VISITNUM	VISIT
YAMAA-01	PP	1	MRCMAX	Metabolite Ratio Cmax	0.4	RATIO	1	PERIOD 1
YAMAA-01	PP	2	MRCMAX	Metabolite Ratio Cmax	0.5	RATIO	2	PERIOD 2
YAMAA-01	PP	1	MRCMAX	Metabolite Ratio Cmax	0.6	RATIO	1	PERIOD 1
YAMAA-01	PP	2	MRCMAX	Metabolite Ratio Cmax	0	RATIO	2	PERIOD 2

Unexpected Output

Under reviewNo artifact is produced

No fixture files in this directory.

Shaded columns are derived— = missing value

Human review and assessment

Dataset Name	Variable Name	Variable Label	Type	Computational Method	Origin
ADSL	STUDYID	Study Identifier	character	Copy value from DM.STUDYID.	Predecessor
ADSL	USUBJID	Unique Subject Identifier	character	Copy value from DM.USUBJID.	Predecessor
ADSL	MRGM	Geometric Mean Metabolite Ratio Cmax	numeric	Aggregate GEOMEAN(PP.PPSTRESN) over records matching on STUDYID and USUBJID, where PP.PPTESTCD = 'MRCMAX'.	Derived

Version	Date	Author	Description
1.7	2026-10-20	yamaa docs build	Read a specification in the 1.0 surface as its 0.5 counterpart; state that missing values stay missing where a mapping declares no missing handler, and leave out a declared missing answer that restates a handler's default
1.6	2026-10-17	yamaa docs build	Describe collected date parts written as ISO 8601 date text and times of day counted in seconds since midnight
1.5	2026-10-17	yamaa docs build	Name collected precision in words instead of letter codes; state imputation fallbacks, minimum source precision, bounds ignored outside the collected period, and invalid text with no handler as an error; describe date and datetime shifts, imputation flag codes, and partial text giving no date part
1.4	2026-10-15	yamaa docs build	State datetime precision codes through the hour and minute, and complete only the time parts a collected datetime lacks
1.3	2026-09-29	yamaa docs build	Show study codelist names, external dictionaries, and ISO 8601 formats in variable rows; list the document's codelist values and metadata
1.2	2026-09-29	yamaa docs build	Classify ADaM copies from a declared input as Predecessor and recodes as Derived, and SDTM recodes of locally derived columns as Derived
1.1	2026-09-29	yamaa docs build	Resolve lookups into the donor record they select, and give every derivation shape a plain-language rule instead of its raw YAML
1.0		yamaa docs build	Initial generation from spec.yaml

How to fix

input/dm.csv

input/pp.csv

Your review

Mapping spec