Skip to content

665912ed00000192 GPI anchor biosynthetic process GO 0006506

Based on my review of the GO-CAM model "GPI anchor biosynthetic process (GO:0006506)" with ID gomodel:665912ed00000192, here's my assessment:

General Review

This model represents the GPI (glycosylphosphatidylinositol) anchor biosynthetic process in Schizosaccharomyces pombe (fission yeast), which is a complex multi-step pathway involving many enzymes that work sequentially to synthesize GPI anchors for proteins.

Strengths of the Model

  1. Comprehensive representation: The model includes many of the key enzymes involved in GPI anchor biosynthesis, with appropriate molecular functions assigned.

  2. Proper cellular locations: The activities are correctly localized to appropriate cellular components (endoplasmic reticulum membrane, ER, GPI-anchor transamidase complex, etc.).

  3. Causal connections: Most activities are properly connected with causal relationships using the appropriate "provides input for" (RO:0002413) predicate to show the sequential steps in the pathway.

  4. Evidence: The model uses appropriate evidence codes, with a mix of experimental evidence, sequence similarity, and orthology-based annotations.

Issues Identified

  1. Duplicate causal relationships: There are several instances where the same causal relationship is represented twice between the same activities. For example:
  2. In activity gomodel:665912ed00000192/665912ed00000210, there are two identical causal associations to activity 665912ed00000232
  3. In activity gomodel:665912ed00000192/665912ed00000396, there are duplicate causal associations to activity 665912ed00000193

  4. Several other activities have this same issue throughout the model

  5. Complex representation: The model comment mentions "deal with 2 complexes once in complex portal," but it's not clear if this has been addressed. Based on the GO-CAM guidelines for representing complexes, these should be properly annotated according to whether the subunit carrying the activity is known or not.

  6. Missing molecular functions: Some activities have GO:0003674 (molecular function) assigned, which is the root term for all molecular functions, indicating that the specific function is not known. While this is acceptable when the function is truly unknown, it should be reviewed to ensure more specific terms aren't available.

  7. Unused inputs and outputs: Some activities like gomodel:665912ed00000192/665912ed00000427 have inputs defined (CHEBI:16749) but this input is listed twice. Activity gomodel:665912ed00000192/66c7d41500001900 has an output defined but isn't clearly connected to the rest of the pathway.

  8. Lack of regulation clarity: The model shows several regulatory relationships (using RO:0002629 "directly positively regulates") but the mechanism of this regulation isn't always clear.

Recommendations

  1. Remove duplicate causal associations: Clean up the model by removing duplicate causal relationships between the same activities.

  2. Clarify complex representation: According to the best practices document, if subunits carrying specific activities are known, they should be represented directly. If not, use the complex ID. The comment about handling complexes should be addressed.

  3. Review molecular function annotations: For activities with GO:0003674, consider if more specific molecular function terms could be applied based on recent literature.

  4. Connect isolated activities: Ensure all activities, especially those with defined inputs/outputs like activity 66c7d41500001900 (pis1), are properly connected to the pathway.

  5. Verify evidence for regulatory relationships: Ensure that all regulatory relationships have appropriate evidence.

  6. Update comment: Once the issue with the complexes is resolved, the comment "deal with 2 complexes once in complex portal" should be updated or removed to reflect the current state of the model.

Summary

This is a comprehensive GO-CAM model of the GPI anchor biosynthetic process in S. pombe. While it accurately represents most aspects of the pathway, there are issues with duplicate causal relationships and some potential improvements in the representation of molecular functions and complexes. Addressing these issues would make the model more accurate and easier to interpret.