CG6051

Summary: IBA annotation supported by human ortholog ZFYVE28/LST2, which acts as a negative regulator of EGFR signaling downstream of mTORC1. Strong evidence from biochemical and functional studies in mammalian systems.

Reason: This is a well-supported IBA annotation. The human ortholog ZFYVE28/LST2 has been demonstrated to negatively regulate EGFR signaling as a substrate of mTORC1, providing a feedback mechanism, and CG6051/Q9VB70 is explicitly listed as the Drosophila LST2 homolog in the Battaglioni 2024 cross-species analysis. The falcon deep research corroborates this family-level function but adds an important caveat: there is NO direct Drosophila experiment testing EGFR regulation by CG6051/Q9VB70, and worm/human genetics frame lst-2-family proteins as signaling MODULATORS rather than core pathway components. The annotation is retained as ACCEPT because it captures the best-supported, family-conserved function, but the inference-based (rather than direct fly) nature of the evidence should be borne in mind.

Summary: IBA annotation based on FYVE domain that specifically binds PI3P on early endosomes. FYVE domains are well-characterized endosomal targeting modules. Human ortholog LST2 localizes to endosomes.

Reason: The FYVE zinc finger domain present in CG6051 (residues 909-969) is a well-established PI3P-binding module that specifically targets proteins to early endosomal membranes. This is a defining characteristic of all FYVE domain-containing proteins. The human ortholog ZFYVE28/LST2 has been experimentally shown to localize to endosomes via its FYVE domain. The domain architecture and conservation strongly support this cellular component annotation.

Summary: Generic metal ion binding annotation based on FYVE zinc finger domain. While technically correct, zinc ion binding (GO:0008270) is more specific and informative.

Reason: This annotation is correct but too general. The FYVE domain specifically coordinates zinc ions through conserved cysteine and histidine residues. The more specific term "zinc ion binding" (GO:0008270) already exists in the annotations and better represents the molecular function. Metal ion binding is a parent term that adds little information.

Proposed replacements: zinc ion binding

Summary: Incorrect annotation. CG6051/LST2 is NOT a phosphatase. It binds PI3P via its FYVE domain but does not have phosphatase activity. This is a misannotation likely from automated prediction.

Reason: This annotation is fundamentally incorrect. CG6051 contains a FYVE domain that BINDS phosphatidylinositol-3-phosphate (PI3P) but does not possess phosphatase activity. The protein lacks any phosphatase catalytic domains. The human ortholog ZFYVE28/LST2 is well-characterized as an adaptor/scaffold protein that is phosphorylated BY kinases (mTORC1), not as a phosphatase itself. This appears to be an erroneous automated prediction conflating PI3P binding with PI3P phosphatase activity.

Summary: Incorrect annotation. CG6051/LST2 is a phosphorylation substrate, not a phosphatase. It is phosphorylated BY mTORC1, not a phosphatase itself.

Reason: This annotation is incorrect. CG6051/LST2 does not have phosphatase activity. Instead, it is a SUBSTRATE of phosphorylation - the protein is phosphorylated by mTORC1 kinase at multiple serine residues (S549, S550, S810 in Drosophila). The protein lacks any phosphatase catalytic domains. This is another erroneous automated prediction that misinterprets the protein as having enzymatic activity when it is actually an adaptor/regulatory protein.

Summary: Partial annotation. While CG6051/LST2 is cytosolic, it more specifically localizes to early endosomes via its FYVE domain. Cytosol alone is too broad.

Reason: This annotation is not incorrect but incomplete. CG6051 is indeed a cytosolic protein (not transmembrane or secreted), but it specifically associates with early endosomal membranes through its FYVE domain binding to PI3P. The cytosol annotation captures the general localization but misses the functionally important endosomal membrane association. The early endosome membrane annotation (GO:0031901) is more informative for understanding protein function.

Summary: Correct annotation. The FYVE domain coordinates two zinc ions through conserved cysteine and histidine residues. This is a defining structural feature required for domain folding and PI3P binding.

Reason: This annotation is well-supported. The FYVE zinc finger domain in CG6051 (residues 909-969) contains eight conserved cysteine/histidine residues that coordinate two zinc ions in a cross-braced topology. UniProt identifies specific zinc-binding residues at positions 915, 918, 931, 934, 939, 942, 961, and 964. Zinc coordination is essential for FYVE domain folding and function. This is a core molecular function of the protein.

Summary: Too general. While CG6051 associates with membranes via its FYVE domain, the specific annotation "early endosome membrane" (GO:0031901) is more accurate and informative.

Reason: This annotation is too vague. CG6051 does associate with membranes, but specifically with early endosomal membranes through PI3P binding via its FYVE domain. The generic "membrane" term provides little functional information. The more specific "early endosome membrane" (GO:0031901) annotation already exists and better captures the protein localization.

Proposed replacements: early endosome membrane

Summary: Too vague. While CG6051/LST2 likely binds kinases like mTORC1 and may interact with EGFR, "enzyme binding" is uninformative. More specific terms like "protein kinase binding" would be better.

Reason: This annotation is likely correct but too general to be useful. The human ortholog LST2 is known to be phosphorylated by mTORC1 kinase, suggesting it binds this enzyme. It may also interact with EGFR or other signaling proteins. However, "enzyme binding" is so broad it provides minimal functional insight. Without specific experimental evidence in Drosophila for which enzymes are bound, this annotation adds little value. More specific terms like "protein kinase binding" (GO:0019901) would be more appropriate if supported by evidence.

Summary: Incorrect annotation. CG6051/LST2 binds phospholipids (PI3P) but is not involved in their biosynthesis. This appears to be a misinterpretation of its lipid-binding function.

Reason: This annotation is incorrect. CG6051/LST2 does not participate in glycerophospholipid biosynthesis. The protein binds to phosphatidylinositol-3-phosphate (a phospholipid) via its FYVE domain but has no biosynthetic activity. There are no lipid biosynthesis domains in the protein. This appears to be an erroneous automated prediction that confused lipid binding with lipid biosynthesis. The protein functions in signaling and endosomal trafficking, not in lipid metabolism.

Summary: Incorrect annotation. CG6051/LST2 binds PI3P but does not dephosphorylate it. The protein lacks phosphatase domains and catalytic activity.

Reason: This annotation is incorrect. CG6051/LST2 does not catalyze phosphatidylinositol dephosphorylation. The protein binds PI3P through its FYVE domain but has no phosphatase activity. This is another instance of automated prediction incorrectly inferring enzymatic activity from lipid-binding capability. The protein is an adaptor/scaffold in signaling pathways, not a metabolic enzyme. All phosphatase-related annotations for this protein are erroneous.

Summary: Incorrect annotation. CG6051/LST2 is not a phosphatase. This is another erroneous automated prediction assigning phosphatase activity where none exists.

Reason: This annotation is completely incorrect. CG6051/LST2 has no phosphatase activity whatsoever. The protein contains only a FYVE zinc finger domain for PI3P binding and lacks any phosphatase catalytic domains. This specific phosphatase activity annotation is particularly problematic as it suggests a very specific enzymatic function that the protein absolutely does not possess. The human ortholog is well-characterized as a signaling adaptor, not an enzyme. This is a clear case of over-annotation by automated systems.

Summary: Correct annotation. CG6051/LST2 functions as an adaptor/scaffold protein in the mTORC1-EGFR signaling axis, connecting kinase signaling to receptor regulation.

Reason: This annotation is well-supported. The human ortholog ZFYVE28/LST2 is characterized as an adaptor/scaffold protein that links mTORC1 signaling to EGFR regulation. It serves as a molecular bridge, being phosphorylated by mTORC1 and then mediating negative feedback on EGFR. The protein lacks enzymatic activity but functions to organize signaling complexes on endosomes. This molecular adaptor activity is a core function of the protein and consistent with its domain architecture (FYVE domain for membrane localization, extensive unstructured regions for protein interactions).

Summary: Unsupported annotation. While FYVE proteins can be involved in cytokinesis, there is no specific evidence linking CG6051/LST2 to midbody abscission.

Reason: This annotation lacks supporting evidence. While some FYVE domain proteins participate in cytokinesis and midbody abscission through endosomal trafficking roles, there is no specific evidence that CG6051/LST2 is involved in this process. The protein was tested in fertility screens but showed no significant phenotype. The human ortholog is characterized for its role in growth factor signaling, not cytokinesis. This appears to be an over-reaching automated prediction based on weak domain similarity to other FYVE proteins that do have cytokinesis roles. Without experimental support, this annotation should be removed.

Summary: cytoplasm identified from core_functions analysis

Reason: This cellular component term reflects CG6051's cytoplasmic localization as a peripheral membrane-associated protein that lacks signal peptides or transmembrane domains.

Summary: endocytosis identified from core_functions analysis

Reason: This biological process term captures CG6051's role in endocytic trafficking through its FYVE domain targeting to PI3P-enriched endosomal compartments.

Summary: vesicle organization identified from core_functions analysis

Reason: This biological process term reflects CG6051's role in organizing endosomal vesicles through membrane attachment and receptor trafficking mechanisms.

Summary: phosphatidylinositol-3-phosphate binding identified from core_functions analysis

Reason: This molecular function term captures the specific PI3P-binding activity of CG6051's FYVE domain essential for endosomal membrane targeting and signaling function.