RCI pipeline — Seed Longevity Lab

Preferred format: one row per completed germination test — taxon, accession, storage condition, storage age (days / months / years, or CollectionDate + TestStartDate), number sown/tested and number germinated. Multiple distinct tests may share a storage age. Percent is recomputed from counts; the site never assumes a sample size. Mark Initial tests with is_initial (legacy storage=Initial still recognized). Substrate / treatment / drying / test ID are optional metadata. Columns are matched by name, any order — legacy monitoring exports parse unchanged.

family,taxon,accession,is_initial,storage_condition,substrate,germination_treatment,storage_age_years,tested,germinated
Malvaceae,Abutilon demo,D-001,TRUE,Initial,paper,untreated,0,50,42
Malvaceae,Abutilon demo,D-001,FALSE,-18C,paper,untreated,2,50,40
Malvaceae,Abutilon demo,D-001,FALSE,5C,paper,untreated,2,50,38
Malvaceae,Abutilon demo,D-001,FALSE,-18C,paper,untreated,5,50,36
Malvaceae,Abutilon demo,D-001,FALSE,5C,paper,untreated,5,50,33
Malvaceae,Abutilon demo,D-001,FALSE,-18C,paper,untreated,12,50,33
Malvaceae,Abutilon demo,D-001,FALSE,5C,paper,untreated,12,50,20
Malvaceae,Abutilon demo,D-002,TRUE,Initial,paper,untreated,0,50,30
Malvaceae,Abutilon demo,D-002,FALSE,-18C,paper,untreated,5,50,26
Malvaceae,Abutilon demo,D-002,FALSE,-18C,paper,untreated,10,50,23
Campanulaceae,Cyanea demo,D-014,TRUE,Initial,agar,GA3,0,40,22
Campanulaceae,Cyanea demo,D-014,FALSE,5C,agar,GA3,3,40,20
Campanulaceae,Cyanea demo,D-014,FALSE,5C,agar,GA3,8,40,17
Campanulaceae,Cyanea demo,D-014,FALSE,-18C,agar,GA3,3,40,8
Campanulaceae,Cyanea demo,D-014,FALSE,-18C,agar,GA3,8,40,3

info — 3 row(s) used legacy storage="Initial" — mapped to is_initial=TRUE.

15 completed tests parsed. Your data never leaves the browser. Storage ages keep their full precision (a 1,740-day test is 4.76 y, never rounded to 5), and the baseline is the highest observed germination in the early 5-year window — verified against the handoff's synthetic fixtures and the real 29,020-test dataset.

The analysis will: set an early-window baseline per accession → RCI P70 = 70% of it → find each accession × storage's latest passing age (max-at-age rule, RCI branch only) → attach GLM + Wilson support → roll up to species intervals — Overall observed RCI and Observed RCI at recommended storage reported separately → flag freeze sensitivity → publication sheet in censoring-honest notation, with an RCI management plot for every evaluable accession.