Horse ALDH7A1 (A0A9L0R3Y5): evolutionary grounding and residue conservation

Script: compare_sequences.py (Biopython global alignment, BLOSUM62, gap open -10,
extend -0.5, free end gaps). Raw output: results.json. Run date 2026-10-20.

1. Evolutionary placement

2. Pairwise identity to the PAINT seed orthologs

Comparison Ortholog length Aligned columns Identical Identity
horse A0A9L0R3Y5 vs human P49419 539 539 516 95.7%
horse A0A9L0R3Y5 vs mouse Q9DBF1 539 539 468 86.8%
horse A0A9L0R3Y5 vs horse A0A9L0RRL6 521 521 521 100.0%

Every residue of human P49419 aligns with no internal gap; human residue 1 aligns to horse
residue 28. The other horse translation, A0A9L0RRL6, is identical to A0A9L0R3Y5 over all 521 of its
residues and lacks 18 residues that A0A9L0R3Y5 retains.

3. Catalytic and ligand-binding residues

All 14 single-residue ACT_SITE, BINDING and SITE positions annotated (with experimental evidence)
in human P49419 are conserved in horse A0A9L0R3Y5. Horse numbering = human + 27.

Human position Role (human UniProt) Human Horse position Horse
296 Proton acceptor (active site) E 323 E
330 Nucleophile (active site) C 357 C
195 Transition state stabilizer N 222 N
192, 194, 218, 259, 275, 297, 427 NAD(+) binding T, F, K, T, S, L, E 219, 221, 245, 286, 302, 324, 454 T, F, K, T, S, L, E
331, 489, 490 (S)-2-amino-6-oxohexanoate (AASA) binding T, G, A 358, 516, 517 T, G, A

Result: all_sites_conserved = true. The horse protein retains the full catalytic machinery and
the substrate-binding residues for alpha-aminoadipic semialdehyde. Glu427 (human), described as the
hot-spot residue for pyridoxine-dependent epilepsy mutations (PMID:31652343), is Glu454 in horse.

4. N-terminus

A0A9L0R3Y5 begins with a 27-residue segment, MGCPGRGAGLYFRGSSGRRRVVRPGLS, before the residue
aligned to human Met1. Neither human P49419 nor mouse Q9DBF1 has it. The human 1..26 mitochondrial
transit peptide (ECO:0000255 prediction) aligns to horse 28..53
(MWRVGRQLCVQAAGGIRLSGLWNRPA vs human MWRLPRALCVHAAKTSKLSGPWSRPA). Whether horse translation starts
at the upstream Met1 or at the Met aligned to human Met1 is not established by any data here. Nor
is whether the arginine-rich extension alters mitochondrial import. Mitochondrial localization of
horse ALDH7A1 is therefore inferred, not observed.

Limitations