Ang2 ISO Source Trace Results

Command used:

python3 genes/mouse/Ang2/Ang2-bioinformatics/trace_iso_sources.py

Summary

• The local Ang2-goa.tsv file contains 46 ISO annotations for mouse Ang2.
• Every one of those 46 ISO rows is transferred through GO_REF:0000119 from the same source entity, UniProtKB:P03950 (human angiogenin/ANG).
• The current QuickGO trace divides those source terms into three buckets:
• 34 with current direct experimental support on human ANG.
• 4 with only inferred support on human ANG (GO:0019731, GO:0045087, GO:0050830, GO:0061844).
• 8 with no current matching human ANG source term (GO:0001878, GO:0002181, GO:0006898, GO:0005694, GO:0015629, GO:0019732, GO:0033554, GO:0036416).
• 9 ISO terms overlap a non-ISO mouse Ang2 annotation for the same GO term in the local GOA export, so those transfers are redundant even before biological review.

Biological Interpretation

• Mouse Ang2 is Angrp / angiogenin-related protein, a paralog of canonical mouse/human angiogenin rather than a simple one-to-one functional equivalent. The original biochemical characterization found that mouse Ang is angiogenic but Angrp is not, while Angrp still has appreciable ribonucleolytic activity toward tRNA and related substrates PMID:8633065.
• The later structural study reinforced paralog divergence and showed that mAng-2 is almost 80% active compared to the human form in comparative tRNA cleavage assays, while also noting that mAng-2 is not angiogenic among the murine paralogues [PMID:23170778 "mAng-2 is almost 80% active compared to the human form" ; "Experiments have shown that mAng, mAng-3 and mAng-4 have angiogenic activities comparable to that of hAng, whereas mAng-2 is not angiogenic"].
• Because of that divergence, even ISO terms with real experimental support on human ANG are unsafe to accept automatically for mouse Ang2 when they concern angiogenesis, receptor binding, signaling, stress-granule biology, nuclear trafficking, or immune effector roles.

Review Implications

• GO:0004540 and GO:0004549 are the main ISO terms that remain biologically plausible for Ang2, because mouse Ang2 has direct biochemical evidence for RNase/tRNA cleavage activity PMID:8633065 PMID:23170778.
• Immune/antimicrobial ISO terms are particularly weak because their current human ANG source is itself inferred rather than experimental.
• ISO terms with no current human source term are best treated as stale transfers.
• The non-ISO experimental block in the local GOA file also contains clear homonym errors:
• PMID:17670746 is about angiopoietin-2, not angiogenin-2 PMID:17670746.
• PMID:25313067 and PMID:28096417 are about angiotensin II (AngII) signaling and WNK4, not Ang2/angiogenin-related protein PMID:25313067 PMID:28096417.
• PMID:29065170 is a ZNF418 cardiac hypertrophy paper that uses Ang II as a stimulus, not an Ang2 gene study PMID:29065170.