Annotation Conf. Call September 24th, 2013
Annotation extension exercises
N.B. have replaced example from the one discussed on the call since it was an invalid annotation.
Paper 1: PMID:12479811 (Title: VASA localization requires the SPRY-domain and SOCS-box containing protein, GUSTAVUS.)
http://www.ncbi.nlm.nih.gov/pubmed/12479811
Annotation Extension to be discussed for the cytoplasm annotation for VAS.
DB (Col 2) | GO ID (Col 5) | ev.code | Reference (Col 6) | Extension (Col 16) |
---|---|---|---|---|
vas, P09052 | GO:0005737 (cytoplasm) | IDA | PMID:12479811 | part_of: CL:0000026 (invertebrate nurse cell) |
Evidence is from Figure 2 and text on page 868: "To confirm that VAS, GUS, and EXU are at times associated within the same cytoplasmic particles, we examined the localization of these three proteins at an ultrastructural level. By simul- taneously staining the same sections for VAS, GUS, and EXU, we found them to colocalize in electron-dense structures within the nurse cell cytoplasm (Figure 2I)."
Paper 2: PMID 9799565 (Title: Role of human CYP4F2 in hepatic catabolism of the proinflammatory agent leukotriene B4.)
http://www.ncbi.nlm.nih.gov/pubmed/9799565 File:1-s2.0-S0003986198908803-main(1).pdf
Annotation Extension to be discussed for the leukotriene-B4 20-monooxygenase activity annotation for CYP4F2.
DB (Col 2) | GO ID (Col 5) | ev.code | Reference (Col 6) | Extension (Col 16) |
---|---|---|---|---|
CYP4F2, P78329 | GO:0050051 (leukotriene-B4 20-monooxygenase activity) | IDA | PMID:9799565 | occurs_in: UBERON:0002107 (liver) |
Evidence is from Table III
Paper 3: PMID 23437011 - This is a long paper, so only a few annotations will be discussed. (Title: The Caenorhabditis elegans JNK signaling pathway activates expression of stress response genes by derepressing the Fos/HDAC repressor complex.)
http://www.ncbi.nlm.nih.gov/pubmed/23437011
Annotation Extension to be discussed for KGB-1 phosphorylation of FOS-1
DB (Col 2) | GO ID (Col 5) | ev.code | Reference (Col 6) | Extension (Col 16) |
---|---|---|---|---|
WB:WP:CE29466 or UniProtKB:O44408 | GO:0004674 (protein serine/threonine kinase activity) | IDA | PMID:23437011 | has_direct_input: WB:WP:CE27375 or UniProtKB:G5EDK8 (FOS-1B in paper, the 331 amino acid isoform in WB and UniProtKB) |
Relevant text (not inclusive, i.e., there are other statements that also support the annotation):
As FOS-1A has previously been characterized as a regulator of anchor-cell invasion during nematode development [17], we focused our investigations on the FOS-1B form (hereafter referred to as FOS-1).
Figure 1. FOS-1 is phosphorylated by KGB-1.
We further generated three FOS-1 mutants that individually changed Thr-304, Thr-316, or Thr-318 to Ala and found that the FOS-1(T304A) mutation exhibited decreased phosphorylation by KGB-1 (Figure 1D, line 3 and Figure S2). These results suggest that T304 is a major site of phosphorylation.
To confirm that KGB-1 phosphorylates FOS-1 at the Thr-304 residue, we generated anti-phospho-FOS-1 antibodies that specifically recognize FOS-1 phosphorylated at Thr-304. Transfection with active KGB-1, but not with the kinase-negative mutant KGB-1 KN, resulted in strong reactivity of FOS-1 with this antibody (Figure 1D, lanes 1, 2). In contrast, we found that the FOS-1 (T304A) mutated form could not be detected by this antibody (Figure 1D, lane 3), confirming that it was specific for FOS-1 phosphorylated at Thr-304.
Note: The phosphorylation experiments reportedly rely on an activated form of KGB-1 (activated by another protein kinase, MEK-1) but there doesn't seem to be a control in the paper that definitively shows this. Therefore, I haven't added any additional information about the role of MEK-1. We could discuss this on the call.
Annotation Extension to be discussed for KGB-1 regulation of FOS-1 oligomerization
DB (Col 2) | GO ID (Col 5) | ev.code | Reference (Col 6) | Extension (Col 16) |
---|---|---|---|---|
WB:WP:CE29466 or UniProtKB:O44408 | GO:0032463 (negative regulation of protein homooligomerization) | IDA | PMID:23437011 | has_regulation_target: WB:WP:CE27375 or UniProtKB:G5EDK8 (FOS-1B in paper, the 331 amino acid isoform in WB and UniProtKB) |
Relevant text:
Indeed, GFP-FLAG-FOS-1 readily co-immunoprecipitated with T7-FOS-1 (Figure 1E, lanes 1, 2), indicating that the two proteins oligomerized, presumably as dimers.
Co-expression of active but not inactive KGB-1 resulted in reduced co-immunoprecipitation of T7-FOS-1 with GFP-FLAG-FOS-1 (Figure 1E, lanes 3, 4).
Annotation Extension to be discussed for FOS-1 regulation of kreg-1 transcription
DB (Col 2) | GO ID (Col 5) | ev.code | Reference (Col 6) | Extension (Col 16) |
---|---|---|---|---|
WB:WBGene00001345 or UniProtKB:G5EDK8, UniProtKB:G5ECG2 | GO:0000122 (negative regulation of transcription from RNA polymerase II promoter) | IMP | PMID:23437011 | has_regulation_target: WB:WBGene00018725 or UniProtKB:Q20689 (kreg-1) |
Relevant text:
Treatment with fos-1 RNAi markedly increased intestinal Pkreg-1::venus expression even in the absence of Cu2+ (Figure 4).
The effect of fos-1 RNAi on expression of kreg-1 and kreg-2 was further confirmed by qRT-PCR (Figure S6).
Discussion
- Pipe vs comma in col-16: Pipe is like making a separate annotation. Pipe can be interpreted as meaning "or" and comma meaning "and"
- with_product_of relationship: when you mention a gene name as a substrate for a enzyme, would you use with_product_of to indicate that the product is the substrate? Not necessary. We are not being specific for With column.
- question about Mek1 phosphorylating KGB-1: Worm paper doesn't have evidence that Mek1 phosphorylates KGB-1, although they mention it. So can't capture that detail from this paper. But the yeast paper discussed last month had evidence for two proteins being kinased in the same paper.
- regulation_target: Do we have to say has input gene, has output RNA? For now we will just put the gene_id in col-16 for has_regulaiton_target
- User_label in the relationship ontology file. We can use that to make sure the users see meaningful stuff (e.g. has_direct_input vs has_substrate) We will talk about it at Barharbor. Pombase has a mapping between relationships and meaningful labels. Midori will look into adding these user labels to the relationship ontology file