# Pre-registered OpenScientist batch for the TreeGrafter evaluation.
# Frozen 2026-10-14 before any report was read. Each entry is one blinded
# function hypothesis ("GENE has TERM"), run with
# templates/treegrafter_function_hypothesis.md in --as-function-hypothesis
# mode. review_action_at_selection is the gene review's action on that
# TreeGrafter (GO_REF:0000118) row at commit 1b7f8f330; compare reports
# against these, not against later actions.
# record_slug (one entry, DDO) selects the review row when the term alone is
# ambiguous: DDO has two GO:0019478 rows, and the run tests the TreeGrafter one.
batch: openscientist-controls-2026-10-14
selected_at_commit: 1b7f8f330
template: templates/treegrafter_function_hypothesis.md
groups:
  undecided: TreeGrafter rows the review left UNDECIDED; the run should help decide them.
  accepted_control: TreeGrafter rows that were the only source for the term and were accepted.
  rule_test: Rows that test a propagation rule or a TreeGrafter-versus-other-pipeline call.
hypotheses:
  - {organism: PSEPK, gene: PP_1969, term: "GO:0061798", label: "GTP 3',8'-cyclase activity", group: undecided, review_action_at_selection: UNDECIDED, why: MoaA paralog in PTHR22960, the family with a fusion-seeded root IBD}
  - {organism: PSEPK, gene: PP_2482, term: "GO:0061798", label: "GTP 3',8'-cyclase activity", group: undecided, review_action_at_selection: UNDECIDED, why: second MoaA paralog in PTHR22960}
  - {organism: PSEPK, gene: PP_3571, term: "GO:0008777", label: acetylornithine deacetylase activity, group: undecided, review_action_at_selection: UNDECIDED, why: mixed M20A family (as dapE)}
  - {organism: PSEPK, gene: PP_4594, term: "GO:0004123", label: cystathionine gamma-lyase activity, group: undecided, review_action_at_selection: UNDECIDED, why: UniProt names it cystathionine gamma-synthase}
  - {organism: PSEPK, gene: ptxD, term: "GO:0030267", label: glyoxylate reductase (NADPH) activity, group: undecided, review_action_at_selection: UNDECIDED, why: phosphonate dehydrogenase; trips the EC-contradiction rule}
  - {organism: PSEPK, gene: gabD-II, term: "GO:0004777", label: succinate-semialdehyde dehydrogenase (NAD+) activity, group: undecided, review_action_at_selection: UNDECIDED, why: trips the EC-contradiction rule}
  - {organism: SACEN, gene: eryCIV, term: "GO:0008483", label: transaminase activity, group: undecided, review_action_at_selection: UNDECIDED, why: deoxysugar biosynthesis enzyme; generic transaminase term}
  - {organism: DOROP, gene: ADAR2, term: "GO:0003726", label: double-stranded RNA adenosine deaminase activity, group: accepted_control, review_action_at_selection: ACCEPT, why: sole source of the defining activity}
  - {organism: HETGA, gene: Cgas, term: "GO:0061501", label: "2',3'-cyclic GMP-AMP synthase activity", group: accepted_control, review_action_at_selection: ACCEPT, why: sole source of the defining activity}
  - {organism: PSEPK, gene: relA, term: "GO:0008728", label: GTP diphosphokinase activity, group: accepted_control, review_action_at_selection: ACCEPT, why: sole source of the defining activity}
  - {organism: 9AVES, gene: TAS1R3, term: "GO:0033041", label: sweet taste receptor activity, group: accepted_control, review_action_at_selection: ACCEPT, why: sole source; hummingbird T1R3}
  - {organism: PSEPK, gene: fadB, term: "GO:0016509", label: "long-chain (3S)-3-hydroxyacyl-CoA dehydrogenase (NAD+) activity", group: rule_test, review_action_at_selection: ACCEPT, why: accepted although it trips the EC-contradiction rule}
  - {organism: PSEPK, gene: metZ, term: "GO:0016846", label: carbon-sulfur lyase activity, group: rule_test, review_action_at_selection: ACCEPT, why: accepted although it trips the EC-contradiction rule}
  - {organism: OCTVU, gene: DDO, term: "GO:0019478", label: D-amino acid catabolic process, group: rule_test, review_action_at_selection: ACCEPT, record_slug: existing-go-0019478-accept, why: "accepted although the query lies outside the IBD node's clade (Eubacteria); the term also has an IDA row (PMID:7915543)"}
  - {organism: PSEPK, gene: PP_2320, term: "GO:0071972", label: "peptidoglycan L,D-transpeptidase activity", group: rule_test, review_action_at_selection: ACCEPT, why: kept over ARBA carboxypeptidase activity (REMOVE)}
  - {organism: PSEPK, gene: pdxH, term: "GO:0042823", label: "pyridoxal 5'-phosphate biosynthetic process", group: rule_test, review_action_at_selection: ACCEPT, why: kept over InterPro2GO pyridoxine biosynthetic process (REMOVE)}