Batch query: columns to be diplayed

[viomunoz]

The batch query portal is a popular feature in our website. It has been long realised that the Batch Query portal requires improvement of data display.

We assume an allele-based display of the data (each row is an allele split by zygosity and life stage), except when the user request a gene-based display, in which case we can merge row by gene (see ticket #498).

The following data can be displayed, columns following this order:

• Input provided by the user
• Mouse gene symbol
• Mouse gene ID
• Mouse gene Ensembl ID
• IMPC allele symbol
• IMPC allele ID
• Lifestage
• Zygosity
• Background strain name
• Background strain ID
• Production status (latest status, as displayed in the search results pages, either: in production, mice produced, phenotype data available)
• MP term names
• MP term IDs
• Top level MP names
• Top level MP term IDs
• Procedures performed names
• Procedures performed IDs
• Parameters obtained names
• Parameters obtained IDs
• Human gene symbol
• Human gene HGNC ID
• Human gene Ensembl ID

[viomunoz]

We will send e-mail to users to get feedback.

[jmason-ebi]

Can we add lifestage to this list?

[ficolo]

We should discuss the user stories for some of the columns, like, what's the goal of the user for the tested "parameter id" column, phrasing it like:

• As a researcher (member of the consortium) I want to have the possibility of filter my batch query results by parameter id

Sometimes helps to figure out why a feature is important or not important, for instance including the parameter names would make both the table and the TSV output difficult use, the average parameter count for one gene in our dataset is 213 parameters, so the average column in the output will have 213 parameter names, and each one is usually long.

Genes like Zzef1 have 1547 parameters!

[viomunoz]

Having a meeting to discuss this April 8.

[viomunoz]

Based on users' request:

• For MP terms -- can we include p-values?
• Procedures performed and not performed -- allows to determine if a phenotype call was not made due to not being tested or no hit was observed when tested >> we can potentially show successful and not sucessful
• Testing Center -- allows to specifically pull data from a given center
• Human disease

[viomunoz]

Meeting Apr 8 minutes https://docs.google.com/document/d/1fLci_wbe52rxVoMvWLhrsnx6WrXHDJS4VjV0TDqKxBk/edit?usp=sharing