DARS2 notes

2026-09-08 paired review

Selected accession A0A9L0SB67; ordinary UniProt record is TrEMBL and cites genome sequencing/submission rather than a gene-specific functional experiment. Human evidence is summarized in the paired human investigation. Sequence comparison records the exact target gaps and transfer limits. The targeted Europe PMC search found no decisive gene-specific horse functional assay warranting a separate horse Edison investigation; general omics/association hits were not counted as mechanistic validation.

Mitochondrial aspartate-tRNA synthetase must accommodate structurally unusual mitochondrial tRNA(Asp); the 2013 structure compares that accommodation with bacterial AspRS. Cross-aminoacylation of bacterial and mitochondrial tRNA(Asp) is a taxonomic/substrate-structure breadth claim, whereas acceptance of tRNA(Asn) is a different aminoacylation-specificity claim. The human GOA includes the latter, so the original experiments matter. The inspected 2013 full text did not settle that point and the 2005 characterization is abstract-only in the cache. Retaining uncertainty is appropriate. Matrix localization is the primary placement, but additional membrane-associated or extracellular activities cannot be dismissed just from that default.

Sources: PMID:15779907(https://pubmed.ncbi.nlm.nih.gov/15779907/), PMID:23275545(https://pubmed.ncbi.nlm.nih.gov/23275545/).

2026-10-20 — Full re-review, independent of the ProtNLM2 predictions

ProtNLM2 output was not opened or used. Deep research (just deep-research HORSE DARS2 --provider falcon --fallback perplexity-lite) failed: falcon "Failed to authenticate", perplexity-lite HTTP 401 insufficient quota. No deep-research file was written; the review uses the cached human DARS2 papers and the new analyses below.

Identity of the benchmark accession

• A0A9L0SB67 (616 aa) is an alternative isoform of horse DARS2, not the canonical product. All four horse UniProt DARS2 entries come from one Ensembl gene, ENSECAG00000004044 (chr5, 7 protein-coding transcripts). A0A9L0SB67 = ENSECAT00000125808 (16 exons); it differs from ENSECAT00000043786 (A0A3Q2HM61, 652 aa, 17 exons) only by skipping one 108-nt in-frame exon (chr5:9524240-9524347) [file:HORSE/DARS2/DARS2-bioinformatics/RESULTS-rereview.md].
• A0A3Q2HM61 is the gene-centric canonical DARS2 entry in the horse reference proteome (UP000002281) and is 1:1 colinear with human Q6PI48 (643 paired residues, 89.9% identity). The Ensembl canonical (F6STJ2, 668 aa) carries one extra 48-nt exon absent from human. No horse DARS2 entry is Swiss-Prot reviewed.
• The skipped exon removes human-equivalent residues 341-376, at the start of the bacterial-type insertion domain (human 337-470 PMID:23275545), including the acetylated K368. All ATP/aspartate-binding residues (244-247, 266-268, 535, 542, 584-587), the arginine-rich transit region and the OB-fold anticodon-binding domain are retained. Whether the isoform is stable or active is unknown.

Phylogenetic grounding

Unpruned PANTHER 19 PTHR22594 tree: horse DARS2 leaf PTN001475968 and human PTN002503203 share the path PTN000502375 > PTN000502597 (SF5 root) > PTN000502598 > ... > PTN000502607. PTN000502597 carries the PAINT IBDs for aspartate-tRNA ligase activity and aspartyl-tRNA aminoacylation (donors yeast MSD1, E. coli AspS, human DARS2); PTN000502598 carries mitochondrion. PAINT did not propagate human-only aspartate-tRNA(Asn) ligase activity (GO:0050560) or mitochondrial asparaginyl-tRNA aminoacylation (GO:0070145); those are treated as deliberate omissions and not transferred.

Changes to the review

• gene_symbol set to DARS2 (Ensembl/PANTHER symbol for this gene) instead of the accession.
• nucleotide binding: KEEP_AS_NON_CORE -> TOO_GENERAL (entailed by ATP binding / ligase).
• nucleic acid binding: KEEP_AS_NON_CORE -> MODIFY to tRNA binding (OB-fold anticodon domain; conserved tRNA-contacting residues PMID:23275545).
• ATP binding: ACCEPT -> KEEP_AS_NON_CORE (component of the ligase reaction).
• cytoplasm: KEEP_AS_NON_CORE -> MODIFY to mitochondrion (PAINT PTN000502598 level; human mt-AspRS in soluble and membrane mitochondrial fractions PMID:30006346).
• tRNA aminoacylation for protein translation: ACCEPT -> MODIFY to GO:0070146 mitochondrial aspartyl-tRNA aminoacylation. Beware GO:0070145 is the asparaginyl term.
• aminoacyl-tRNA ligase activity and ligase activity: MODIFY to aspartate-tRNA ligase activity (unchanged).
• Core function now names GO:0070146 and mitochondrion; description rewritten to state the isoform status.
• Previous review's supporting quotes from RESULTS.md (pairwise identity) replaced with RESULTS-rereview.md; RESULTS.md kept as a reference.

Relative to review next

Same-gene canonical product A0A3Q2HM61 (652 aa) has no review in the repo (genes/HORSE/DARS2 is keyed to A0A9L0SB67), so the next target is genes/HORSE/DARS2__A0A3Q2HM61. Human DARS2 (Q6PI48) has a full review in genes/human/DARS2 (status DRAFT, no PENDING rows).

Family and modules

• interpro/panther/PTHR22594/PTHR22594-review.yaml lists this accession and Q6PI48 under SF5; placement agrees. Minor issue: the family review's preferred_name/summary describe the family as "Aspartyl/asparaginyl" whereas PANTHER's official name is "ASPARTYL/LYSYL-TRNA SYNTHETASE"; and it lists the exon-skipping isoform rather than the canonical horse product as a representative.
• Module: modules/mitochondrial_trna_charging_and_initiator_formylation.yaml (human DARS2 is a part).

2026-11-06 — tRNA aminoacylation BP obsoletion

GO obsoleted the 42 amino-acid-specific tRNA aminoacylation biological processes,
GO:0070146 mitochondrial aspartyl-tRNA aminoacylation among them, because the amino acid
is already carried by the paired molecular function. Verified in go-edit.obo:
GO:0070146 is is_obsolete: true with replaced_by: GO:0070127, and its comment names
GO:0004815 aspartate-tRNA ligase activity as the molecular function.

This supersedes two decisions recorded above:

• "tRNA aminoacylation for protein translation: ACCEPT -> MODIFY to GO:0070146" is now
  MODIFY to GO:0070127 tRNA aminoacylation for mitochondrial protein translation.
• "Core function now names GO:0070146" is now GO:0070127.

The derivation that produced GO:0070146 — the conjunction of PAINT's aspartyl-tRNA
aminoacylation (PTN000502597) and mitochondrion (PTN000502598) — is still sound about the
facts, but its conclusion is no longer a term: GO:0006422 went in the same batch, so only
the compartment half of the conjunction remains a process. Aspartate specificity stays on
the GO:0004815 rows this review already proposes.

The caution above about GO:0070145 being the asparaginyl term no longer applies: both
terms are obsolete, so the confusion it guarded against can no longer arise.