Skip to content

Config file

Jump to bottom
Kim Rutherford edited this page Jan 24, 2021 · 46 revisions

File location

The config file is in Git: https://github.com/pombase/pombase-config/blob/master/website/pombase_v2_config.json

The following setting are available:

extension_display_names

To appear in the "Target of" part of the gene page the annotation must have a extension relation that matches a "rel_name" in this section. Some items have a "if_descendent_of" setting. If that is present then the term in the annotation must be a descendent of the configured term.

eg.

{
 "rel_name": "has_direct_input",
  "if_descendent_of": "GO:0003824",
  "display_name": "has substrate",
  "reciprocal_display": "substrate of"
},

This will match any annotation where the term is a descendent of "GO:0003824" and has "has_direct_input" as a extension relation.

The string in the "reciprocal_display" field is used in the "Relationship" column of the target of gene page section.

predefined_queries

Preconfigured links to the query builder and "Canned queries" in the query builder are configured here.

The format is:

  "<QUERY_NAME>": {
    "name": "<USER_FRIENDLY_NAME>",
    "constraints": "<QUERY_CONSTRAINTS>"
  }

The <QUERY_NAME> can be any string as long it's unique.

The <USER_FRIENDLY_NAME> is a short title or description of the query that can be shown to users. For canned queries (see canned_query_ids below) this friendly name is shown in the "Canned queries" section of the query builder. For other predefined queries the name is currently not used.

The <QUERY_CONSTRAINTS> is the internal JSON representation of the query. This text is available via a hidden menu to the right of each query.

example:

  "physical_interactors_of_wee1": {
    "constraints":{
      "interactors":{
        "gene_uniquename": "SPCC18B5.03",
        "interaction_type": "physical"
      }
    }
  }

Whitespace outside of the quotes isn't significant and the newlines are optional.

This can be used in a query in the documentation like this. The <QUERY_NAME> from predefined_queries is the predefinedQueryName in the query-link, e.g.

<app-query-link [goToResults]="true" [linkText]="'Physical interactors of wee1'"
    [predefinedQueryName]="'physical_interactors_of_wee1'">
</app-query-link>

See Links to the query builder for more on query links.

canned_query_ids

front_page_panels

A list of settings for the front page panels. The configuration for each panel is chosen randomly from this list on each visit to the front page.

title

The text to include in the title of the panel. Required.

title_link

Optional external link URL.

panel_type

One of "explore", "community" or "spotlight". Required.

head_image

A list of images for the top of this panel. The image will be chosen at random from this list on each visit to the front page. Required'

content

Markdown content for this panel. Required.

link

A link for the bottom of the panel. Optional.

link_label

The text to use for the link, required when link is configured.

show_on_front_page

If true, this panel config is a candidate for the front page. If false the panel is only visible in the archive. Required.

external_gene_references

See also "Adding a new external reference".

Sub-fields of external_gene_references:

ref_type

The string to appear in the first column of the external references table

feature_types

This reference will only appear on gene page with these types. Some be an array with the values "mRNA gene" and/or "pseudogene". It can also be "ALL" so it will appear on all pages including RNA gene pages.

name

The string to appear in column 2.

description

Column 4 contents.

field_name

The field of the gene to substitute into the configured URL.

Possibilities:

  • uniquename: the systematic ID
  • name: the gene name
  • uniprot_identifier
  • orfeome_identifier

and the special case: NCBI_ALL_IDS which will create a string containing name, uniquename and synonyms suitable for a link to PubMed.

url

The URL of the external site. The <<IDENTIFIER>> in the string will be replaced with the value of the field set with the field_name setting.

go_xrf_abbrev

Instead of providing a url you can provide a database abbreviation that matches an entry in GO.xrf_abbs

extension_relation_order

evidence_types

cv_config

There is a configuration item here for each ontology annotation section on the gene pages.

Attributes:

feature_type

This defaults to "gene" and is set to "genotype" for the phenotype annotation sections.

display_name

The title to use at the top of the section. eg. "Single allele phenotype"

filters

The filters item is a list of the configuration for the filters that should shown for this section.

Attributes:

display_name

The name for the filter that is shown to the user.

filter_type

A unique identifier for this filter. eg. "evidence"

detailed_view_only

Should be "true" for filters the should only be shown in the detail (non-summary) view

term_categories

For term type filters this is used to group the annotation by ancestor term. Each category has:

  • a display_name like "Normal process"
  • an ancestors list with the ancestor term ID and the relation to follow to get to the ancestor

There can be multiple ancestors per category. The category will contain all terms that have any of the configured ancestors.

evidence_categories

Each has:

  • a display_name like "Western blot"
  • an evidence_codes list

The evidence codes can be any case. They shouldn't have the word "evidence" at the end. eg. use "Western blot" not "Western blot evidence". That's because we have both versions in Chado (needs fixing) so the web code checks for annotations with and without the " evidence" at the end as the evidence code.

summary_relations_to_hide

summary_gene_relations_to_collect

extra_database_aliases

This map contains aliases for databases that aren't available in GO.xrf_abbs. These aliases are also used for fixing the database names for output in ProteinFeatures.tsv.

The keys should be lower case and the code will then check for any case. eg. for the configuration "amigo": "GO" the code will look for "amigo", "AMIGO" or "AmiGO" and map it to "GO".

term_display_names

A map from term name to a tidier name for display purposes. Currently used for the misc. sequence feature table. See: https://github.com/pombase/website/issues/37#issuecomment-765301311

sequence_feature_page

Configuration for the sequence feature page. See: https://github.com/pombase/website/issues/37

so_types_to_show

The SO type names to show on the sequence feature page.

PomBase is funded by the Wellcome Trust

Clone this wiki locally