Phenotype pages: unclear how number of associations are counted

[viomunoz]

In this Phenotype page: impaired glucose tolerance https://www.mousephenotype.org/data/phenotypes/MP:0005293

In the phenotype page we see:

• 65 significant genes
• Total number of significant genotype-phenotype associations: 98

However, in the downloaded file: There are 104 associations.

• The number of unique genes and alleles is 65
• The difference between 98 and 104 seems to be due to instances in which two different parameters are pointing to the same MP term (when counted once the number of associations is 98).

Example:

Requirements:

• Include a sentence clarifying how the number of significant genotype-phenotype associations is counted.
• Alternatively, or in addition to this, provide a link to a FAQ in the Documentation.
• Add a FAQ reflecting this issue and include in the Documentation, Phenotype pages (section "Data Visualization").

[viomunoz]

Example of explanatory sentence that could be included: "Associations are counted per allele and sex. When more than one parameter is significant, only the most sifgnificant association is displayed (a full list of associations is provided in the downloadable file). Heterozgotes are generally only assessed when homozygotes are lethal."

Note the associations seem to be being counted per allele and sex (e.g. see [limb grasping](https://www.mousephenotype.org/data/phenotypes/MP:0001513; Ap4e1 and Rnf10 have more than one allele). Consider to count per gene and sex.