FER1 and FER2 genotypes displayed on wrong gene pages

[jseager7]

PHI-base 5 has Host Phenotype annotations for the FER1 and FER2 genes of Triticum aestivum. The FER1 genes are included in a multi-allele genotype containing copies of FER1 for multiple genomes (and the same for FER2). However, the gene page for each FER1 gene wrongly shows the genotype that includes FER2:

In the example above, only the FER1 genotype should be displayed. The same should apply for the FER2 gene pages, which wrongly show the FER1 genotype.

Shown below is a table of the FER1 and FER2 genes, with their corresponding UniProtKB accession numbers:

UniProtKB accession number	Gene name
A0A3B5Y122	FER1(a genome)
A0A3B5YY12	FER1(b genome)
A0A3B5ZUX1	FER1(d genome)
A0A3B6HVQ9	FER2(a genome)
A0A3B6ITE6	FER2(b genome)
A0A3B6JHD1	FER2(d genome)

The FER1 genotype should only be shown on the gene pages for the following accessions:

• A0A3B5Y122
• A0A3B5YY12
• A0A3B5ZUX1

The FER2 genotype should only be shown on the gene pages for the following accessions:

• A0A3B6HVQ9
• A0A3B6ITE6
• A0A3B6JHD1

[Molecular-Connections]

Do we need to hardcode the uniprotKB ID for rendering it . If "Yes" do we have any other case also where we need to take care while displaying .

The FER1 genotype should only be shown on the gene pages for the following accessions:

A0A3B5Y122 A0A3B5YY12 A0A3B5ZUX1 The FER2 genotype should only be shown on the gene pages for the following accessions:

A0A3B6HVQ9 A0A3B6ITE6 A0A3B6JHD1

[jseager7]

@Molecular-Connections Sorry, maybe I wasn't making myself clear. What I meant is that after this problem is fixed, then the genotypes should only display on the gene pages for the UniProtKB IDs that I mentioned. I listed the UniProtKB IDs so that you could use them to confirm that the problem is fixed. I didn't mean that the issue itself should be fixed by restricting the genotype to those pages manually.

To clarify: the problem is most likely a data loading issue with the JSON export. For some reason, the PHI-base 5 display is including genotypes on this page that do not contain the UniProtKB ID for the gene in the Entry Summary of the gene page. For example, in the genotype highlighted in the image above, none of the genes in the genotype correspond to the UniProtKB ID A0A3B5YY12.

Here's an example in the JSON export of an allele (part of a genotype) that should be displayed on this page (because the UniProtKB ID is A0A3B5YY12, and that matches the UniProtKB ID in the Entry Summary).

"A0A3B5YY12:925be40b9263f9e8-1": {
    "allele_type": "wild_type",
    "gene": "Triticum aestivum A0A3B5YY12",
    "name": "FER1(b genome)",
    "primary_identifier": "A0A3B5YY12:925be40b9263f9e8-1",
    "synonyms": []
},

While here's an example of an allele that should not be displayed on this page, because the UniProtKB ID does not match:

"A0A3B5ZUX1:925be40b9263f9e8-1": {
  "allele_type": "wild_type",
  "gene": "Triticum aestivum A0A3B5ZUX1",
  "name": "FER1(d genome)",
  "primary_identifier": "A0A3B5ZUX1:925be40b9263f9e8-1",
  "synonyms": []
},

I'm not sure why this problem is happening, but it could probably be fixed by checking every allele in the genotype that is linked to each annotation, and ensuring that the UniProtKB ID for the allele (technically the UniProtKB ID for the gene linked to the allele) matches the UniProtKB ID in the Entry Summary section. So, if the underlying data loading issue can be fixed, then this problem should be fixed as a side-effect of that.

[Molecular-Connections]

Issue has been fixed .

[jseager7]

Confirmed as fixed.