Value-Level Metadata

1library(yamaa)23yamaa_project("submission-project.yaml")$spec("LB")$run()$output

1import yamaa23yamaa.project("submission-project.yaml").spec("LB").run().output

1# Reference solution for the yamaa benchmark sdtm-lb-metadata (R track).2#3# Written from the benchmark's full prompt and its inputs alone, as an agent4# would write it. Harbor's oracle agent runs it to check that the prompt can5# be solved and that the grader scores a correct answer 1.67library(dplyr, warn.conflicts = FALSE)8library(readr)9library(tidyr)1011odm <- read_csv(12  "/app/input/odm.csv",13  col_types = cols(.default = col_character())14)1516key_cols <- c(17  "StudyOID", "SubjectKey", "StudyEventOID", "StudyEventRepeatKey",18  "ItemGroupOID", "ItemGroupRepeatKey"19)2021dates <- odm |>22  filter(ItemOID == "IT.LB.LBDTC") |>23  select(all_of(key_cols), LBDTC = Value)2425tests <- odm |>26  filter(ItemOID %in% c("IT.LB.GLUC", "IT.LB.CREAT")) |>27  left_join(dates, by = key_cols) |>28  mutate(29    LBTESTCD = if_else(ItemOID == "IT.LB.GLUC", "GLUC", "CREAT"),30    LBTEST = if_else(ItemOID == "IT.LB.GLUC", "Glucose", "Creatinine"),31    LBORRES = na_if(Value, ""),32  ) |>33  filter(!is.na(LBORRES)) |>34  mutate(35    DOMAIN = "LB",36    STUDYID = StudyOID,37    USUBJID = SubjectKey,38    LBORRESU = "mg/dL",39    LBSTRESN = suppressWarnings(as.numeric(LBORRES)),40    LBSTRESU = "mg/dL",41  ) |>42  arrange(STUDYID, USUBJID, LBDTC, LBTESTCD) |>43  group_by(STUDYID, USUBJID) |>44  mutate(LBSEQ = row_number()) |>45  ungroup() |>46  select(47    DOMAIN, STUDYID, USUBJID, LBSEQ, LBTESTCD, LBTEST, LBORRES,48    LBORRESU, LBSTRESN, LBSTRESU, LBDTC49  )5051dir.create("/app/output", showWarnings = FALSE)52write_csv(tests, "/app/output/lb.csv", na = "")

1# Reference solution for the yamaa benchmark sdtm-lb-metadata (Python track).2#3# Written from the benchmark's full prompt and its inputs alone, as an agent4# would write it. Harbor's oracle agent runs it to check that the prompt can5# be solved and that the grader scores a correct answer 1.67"""Build the SDTM LB dataset from the ODM extract.89One record per collected glucose or creatinine result. Reads10/app/input/odm.csv and writes /app/output/lb.csv.11"""1213import csv14from collections import OrderedDict1516INPUT_CSV = "/app/input/odm.csv"17OUTPUT_CSV = "/app/output/lb.csv"1819TESTS = OrderedDict(20    [21        ("IT.LB.GLUC", ("GLUC", "Glucose")),22        ("IT.LB.CREAT", ("CREAT", "Creatinine")),23    ]24)25DATE_ITEM = "IT.LB.LBDTC"2627COLUMNS = [28    "DOMAIN",29    "STUDYID",30    "USUBJID",31    "LBSEQ",32    "LBTESTCD",33    "LBTEST",34    "LBORRES",35    "LBORRESU",36    "LBSTRESN",37    "LBSTRESU",38    "LBDTC",39]404142def numeric_form(text):43    """Return the numeric rendering of text, or '' when not a number."""44    try:45        value = float(text)46    except (TypeError, ValueError):47        return ""48    if value == int(value):49        return str(int(value))50    return repr(value)515253def main():54    with open(INPUT_CSV, newline="", encoding="utf-8") as handle:55        rows = list(csv.DictReader(handle))5657    # Group items by subject and visit group (one collection event).58    groups = OrderedDict()59    for row in rows:60        key = (61            row["StudyOID"],62            row["SubjectKey"],63            row["StudyEventOID"],64            row["StudyEventRepeatKey"],65            row["ItemGroupOID"],66            row["ItemGroupRepeatKey"],67        )68        groups.setdefault(key, {})[row["ItemOID"]] = row["Value"]6970    records = []71    for (study_oid, subject_key, *_), items in groups.items():72        lbdtc = items.get(DATE_ITEM, "")73        for item_oid, (testcd, test) in TESTS.items():74            if item_oid not in items:75                continue76            entry = items[item_oid]77            records.append(78                {79                    "DOMAIN": "LB",80                    "STUDYID": study_oid,81                    "USUBJID": subject_key,82                    "LBTESTCD": testcd,83                    "LBTEST": test,84                    "LBORRES": entry,85                    "LBORRESU": "mg/dL",86                    "LBSTRESN": numeric_form(entry),87                    "LBSTRESU": "mg/dL",88                    "LBDTC": lbdtc,89                }90            )9192    # LBSEQ numbers each subject's records from 1 by collection date,93    # then by test code.94    by_subject = {}95    for record in records:96        by_subject.setdefault(record["USUBJID"], []).append(record)97    ordered = []98    for subject in by_subject:99        subject_records = sorted(100            by_subject[subject],101            key=lambda r: (r["LBDTC"], r["LBTESTCD"]),102        )103        for index, record in enumerate(subject_records, start=1):104            record["LBSEQ"] = index105        ordered.extend(subject_records)106107    with open(OUTPUT_CSV, "w", newline="", encoding="utf-8") as handle:108        writer = csv.DictWriter(handle, fieldnames=COLUMNS)109        writer.writeheader()110        for record in ordered:111            writer.writerow({column: record[column] for column in COLUMNS})112113114if __name__ == "__main__":115    main()

1/* opensas reference solution for the yamaa benchmark sdtm-lb-metadata.2   Reads the benchmark inputs and derives the requested output.3   Execute with the existing opensas runtime: opensas /app/output/result.sas. */45data odm;6  length StudyOID MetaDataVersionOID SubjectKey StudyEventOID StudyEventRepeatKey ItemGroupOID ItemGroupRepeatKey ItemOID Value $1024;7  infile "/app/input/odm.csv" dsd dlm="," firstobs=2 truncover lrecl=32767;8  input StudyOID : $char1024. MetaDataVersionOID : $char1024. SubjectKey : $char1024. StudyEventOID : $char1024. StudyEventRepeatKey : $char1024. ItemGroupOID : $char1024. ItemGroupRepeatKey : $char1024. ItemOID : $char1024. Value : $char1024.;9run;1011proc sort data=odm out=ordered; by StudyOID SubjectKey StudyEventOID StudyEventRepeatKey ItemGroupRepeatKey; run;12data dates;13  length LBDTC $1024;14  retain LBDTC;15  set ordered;1617  by StudyOID SubjectKey StudyEventOID StudyEventRepeatKey ItemGroupRepeatKey;18  if first.ItemGroupRepeatKey then do;19    LBDTC='';20  end;21    if ItemOID='IT.LB.LBDTC' then LBDTC=Value;22  if last.ItemGroupRepeatKey then output;23run;24proc sql;25create table joined as select a.*,b.LBDTC from odm a inner join dates b on a.StudyOID=b.StudyOID and a.SubjectKey=b.SubjectKey and a.StudyEventOID=b.StudyEventOID and a.StudyEventRepeatKey=b.StudyEventRepeatKey and a.ItemGroupRepeatKey=b.ItemGroupRepeatKey where a.ItemOID in ('IT.LB.GLUC','IT.LB.CREAT') and not missing(a.Value); quit;26data prepared; set joined; length DOMAIN $2 STUDYID USUBJID LBTESTCD LBTEST LBORRES LBORRESU LBSTRESU LBSTAT $1024;27DOMAIN='LB'; STUDYID=StudyOID; USUBJID=SubjectKey; LBORRES=Value; LBORRESU='mg/dL'; LBSTRESU='mg/dL'; LBSTRESN=input(Value,?? best32.); LBSTAT='';28if ItemOID='IT.LB.CREAT' then do; LBTESTCD='CREAT'; LBTEST='Creatinine'; end; else do; LBTESTCD='GLUC'; LBTEST='Glucose'; end;29if Value='NOT DONE' then do; LBORRES=''; LBORRESU=''; LBSTRESU=''; LBSTRESN=.; LBSTAT='NOT DONE'; end; run;30proc sort data=prepared; by STUDYID USUBJID LBDTC LBTESTCD; run;31data result; set prepared; by STUDYID USUBJID; retain LBSEQ;32  if first.USUBJID then LBSEQ=0;33  LBSEQ+1;34run;3536proc sql;37  create table final as select DOMAIN, STUDYID, USUBJID, LBSEQ, LBTESTCD, LBTEST, LBORRES, LBORRESU, LBSTRESN, LBSTRESU, LBDTC from result;38quit;39proc export data=final outfile="/app/output/lb.csv" dbms=csv replace; run;