Skip to content

Now I have enough information to review the model. Let me provide a comprehensive review of GO-CAM model: gomodel:63a86a8600000489 "TRIM22-NT5C2 regulation of RIGI via ubiquitination (Human)".

Review of GO-CAM Model "TRIM22-NT5C2 regulation of RIGI via ubiquitination (Human)"

This model (gomodel:63a86a8600000489) represents the molecular mechanisms by which TRIM22 and NT5C2 regulate RIG-I (also known as DDX58) through different forms of ubiquitination in human cells, as part of the antiviral innate immune response.

Model Overview

The model includes four activities, representing: 1. NT5C2 (P49902) with ubiquitin protein ligase activity 2. TRIM22 (Q8IYM9) with ubiquitin protein ligase activity 3. RIG-I (O95786) with pattern recognition receptor activity 4. MAVS (Q7Z434) with signaling adaptor activity

The model captures how TRIM22 and NT5C2 differentially regulate RIG-I through distinct types of ubiquitination (K63-linked and K48-linked), affecting the downstream antiviral immune response pathway through MAVS.

Evidence Evaluation

The model is primarily based on the paper by Fei et al. (2022) [PMID:36159777], which provides evidence for the interaction between TRIM22, NT5C2, and RIG-I. The paper demonstrates that TRIM22 promotes K63-linked ubiquitination of RIG-I, while NT5C2 is responsible for K48-linked ubiquitination. This dual regulation affects the RIG-I/NF-κB signaling pathway.

All activities in the model are appropriately supported by ECO evidence codes with proper references, primarily relying on direct assay evidence (ECO:0000314) from the cited publication.

Accuracy Assessment

The model accurately represents the biological mechanism described in the primary paper:

  1. TRIM22 (Q8IYM9) is correctly annotated with ubiquitin protein ligase activity (GO:0061630) and is associated with protein K63-linked ubiquitination (GO:0070534), which activates RIG-I signaling.

  2. NT5C2 (P49902) is correctly annotated with ubiquitin protein ligase activity (GO:0061630) and associated with protein K48-linked ubiquitination (GO:0070936), which negatively regulates RIG-I.

  3. The causal relationships are appropriately captured:

  4. NT5C2 directly negatively regulates RIG-I (RO:0002630)
  5. TRIM22 directly positively regulates RIG-I (RO:0002629)

  6. RIG-I directly positively regulates MAVS (RO:0002629)

  7. RIG-I (O95786) is correctly placed as a pattern recognition receptor (GO:0038187), part of the antiviral innate immune response (GO:0140374).

  8. MAVS (Q7Z434) is appropriately represented with signaling adaptor activity (GO:0035591).

Potential Issues and Recommendations

While the model is generally well-constructed, I identified a few issues that should be addressed:

  1. Redundant causal connection: There are two identical causal connections from TRIM22 to RIG-I (both using RO:0002629). This appears to be a duplication that should be removed.

  2. Missing contextual information: Although the model describes the ubiquitination events well, it would benefit from additional information about the cellular location where this process takes place. Both proteins involved in ubiquitination should have "occurs_in" associations with the cytoplasm (GO:0005737), as correctly indicated for TRIM22 and RIG-I but not explicitly shown for NT5C2.

  3. Biological process annotation: The model could be enhanced by including additional biological process annotations that connect this pathway to interferon production and antiviral defense signaling.

  4. Improved representation of complex: While the model shows separate activities for TRIM22 and NT5C2, the paper suggests they function as a complex. According to the GO-CAM documentation for annotating complexes, when the subunits carrying the molecular activities are known (as in this case), representing the individual proteins as separate activities is appropriate, which this model does correctly.

Compliance with GO-CAM Best Practices

The model follows several GO-CAM best practices: - Each protein is annotated with appropriate molecular functions - Each activity has appropriate "part_of" biological process annotations - Causal relationships between activities use correct predicates - Evidence codes with appropriate reference citations are provided

Specifically for E3 ligase representation, the model follows the guidelines by: - Using "ubiquitin protein ligase activity" (GO:0061630) for both E3 ligases - Specifying the type of ubiquitination (K48-linked vs K63-linked) in the biological process annotation - Using the appropriate causal relations to indicate positive or negative regulation

Summary and Final Assessment

This GO-CAM model accurately represents the dual regulation of RIG-I by TRIM22 and NT5C2 through different types of ubiquitination, as described in the literature. The model is well-structured, uses appropriate GO terms and relationships, and follows the GO-CAM best practices for representing E3 ubiquitin ligase activities.

The minor issues noted above (particularly the redundant causal connection) should be addressed, but they don't significantly impact the overall quality or accuracy of the model. This model effectively captures an important regulatory mechanism in the antiviral innate immune response pathway.