diff --git a/data/genomics/homo_sapiens/riboseq_expression/orf_catalogue/README.md b/data/genomics/homo_sapiens/riboseq_expression/orf_catalogue/README.md
index 8eab5c1e3..5aa5f551f 100644
--- a/data/genomics/homo_sapiens/riboseq_expression/orf_catalogue/README.md
+++ b/data/genomics/homo_sapiens/riboseq_expression/orf_catalogue/README.md
@@ -3,9 +3,16 @@
 Minimal crafted ORF catalogue used as input to the `custom/orfcollapse` module
 test. The real chr20 data does not contain the case orfcollapse handles (the
 same micropeptide encoded at distinct, non-overlapping loci), so this small
-fixture is hand-built to exercise it: two identical-peptide smORFs on opposite
-strands/loci, one canonical CDS and one unique smORF.
+fixture is hand-built to exercise it: two identical-peptide small ORFs on
+opposite strands/loci, one canonical CDS and one unique small ORF.
+
+`orf_class` is positional only — it records where an ORF sits relative to the
+annotated CDS and never encodes length. The `is_smorf` column carries the
+length flag (`aa_length` <= `--smorf-max-aa`), and `orf_type_native` preserves
+each caller's own ORF-type label. The two identical-peptide rows share a class
+deliberately: orfcollapse prefers the more specific class when a peptide cluster
+spans several, so differing classes would change which row survives the fold.
 
 - `cohort.catalogue.{bed12,tsv}`, `cohort.orf_to_gene.tsv`, `cohort.catalogue.mqc.tsv`: orfmerge-shaped catalogue.
 - `cohort.catalogue.aa.fasta`: catalogue peptides (headers carry the `(+)/(-)` suffix bedtools getfasta -nameOnly -s emits).
-- `cohort_cluster.tsv`: MMseqs2 cluster TSV grouping the two identical smORFs.
+- `cohort_cluster.tsv`: MMseqs2 cluster TSV grouping the two identical small ORFs.
diff --git a/data/genomics/homo_sapiens/riboseq_expression/orf_catalogue/cohort.catalogue.mqc.tsv b/data/genomics/homo_sapiens/riboseq_expression/orf_catalogue/cohort.catalogue.mqc.tsv
index 5e6dad808..802251f2e 100644
--- a/data/genomics/homo_sapiens/riboseq_expression/orf_catalogue/cohort.catalogue.mqc.tsv
+++ b/data/genomics/homo_sapiens/riboseq_expression/orf_catalogue/cohort.catalogue.mqc.tsv
@@ -1,3 +1,9 @@
 Class	Count
-smORF	3
 canonical_cds	1
+uORF	2
+uoORF	0
+dORF	1
+doORF	0
+intORF	0
+novel_u	0
+other	0
diff --git a/data/genomics/homo_sapiens/riboseq_expression/orf_catalogue/cohort.catalogue.tsv b/data/genomics/homo_sapiens/riboseq_expression/orf_catalogue/cohort.catalogue.tsv
index eff162213..6d09a2c93 100644
--- a/data/genomics/homo_sapiens/riboseq_expression/orf_catalogue/cohort.catalogue.tsv
+++ b/data/genomics/homo_sapiens/riboseq_expression/orf_catalogue/cohort.catalogue.tsv
@@ -1,5 +1,5 @@
-orf_id	chrom	start	end	strand	gene_id	transcript_id	orf_class	aa_length	called_by_ribotish	called_by_ribocode	called_by_ribotricer	called_by_rpbp	called_by_price	score_ribotish	score_ribocode	score_ribotricer	score_rpbp	score_price	n_samples	samples
-orf_00000001	chr20	100	160	+	g1	t1	smORF	20	1	0	0	0	0	0.01					1	sampleA
-orf_00000002	chr20	9000	9060	-	g5	t5	smORF	20	0	1	0	0	0		0.02				1	sampleB
-orf_00000003	chr20	2000	2900	+	g2	t2	canonical_cds	300	1	0	0	0	0	0.5					1	sampleA
-orf_00000004	chr20	50	125	+	g3	t3	smORF	25	1	0	0	0	0	0.3					1	sampleA
+orf_id	chrom	start	end	strand	gene_id	transcript_id	orf_class	aa_length	called_by_ribotish	called_by_ribocode	called_by_ribotricer	called_by_rpbp	called_by_price	score_ribotish	score_ribocode	score_ribotricer	score_rpbp	score_price	n_samples	samples	orf_type_native	is_smorf
+orf_00000001	chr20	100	160	+	g1	t1	uORF	20	1	0	0	0	0	0.01					1	sampleA	5'UTR	1
+orf_00000002	chr20	9000	9060	-	g5	t5	uORF	20	0	1	0	0	0		0.02				1	sampleB	uORF	1
+orf_00000003	chr20	2000	2900	+	g2	t2	canonical_cds	300	1	0	0	0	0	0.5					1	sampleA	Annotated	0
+orf_00000004	chr20	50	125	+	g3	t3	dORF	25	1	0	0	0	0	0.3					1	sampleA	3'UTR	1