canto.yaml

The default configuration is stored in canto.yaml in the top level directory. Any installation specific settings can be added to canto_deploy.yaml, and will override the defaults. The canto_deploy.yaml will be created automatically by the canto_start start script when it is run with the --initialise option.

The configuration files are YAML format.

name

A one word name for the application. default: Canto

long_name

A longer description of the site. e.g. The SlugBase community annotation tool

database_name

The destination database name. This is used whenever we need to refer to the destination of the annotations while curating.

database_url

The URL of the database that this instance is installed for. e.g. http://www.pombase.org/

instance_front_subtitle

The subtitle to use on the front greeting page.

header_image

A path (relative to root/static) of the logo to put in the header.

message_of_the_day

If set, show this text as a notice at the top of curation session pages.

instance_organism

Canto has two modes: single or multi organism. Single organism mode is activated by setting an "instance organism". That organism will be assumed throughout the interface. Only genes from that single organism can be annotated and the user will not be shown the organism name.

In multi organism mode, genes from any number of organisms can be annotated in each session. In this mode after uploading a list of gene identifiers, the user will be shown the organism name as well as the names, synonyms and products. The organism is shown so the user can confirm that the identifier they gave matched the gene from the right organism. Generally an external gene adaptor should be configured using the instance_organism setting in the multi-organism case.

Example:

instance_organism:
  taxonid: 4896

canto_url

The link for the main Canto web site.

app_version

The software version, automatically updated each release.

schema_version

The version of the schema. This is incremented when the schema changes in an incompatible way.

authentication

Configuration for the Catalyst authentication code. This shouldn't need changing.

view_options

Configuration for the admin view pages.

max_inline_results_length

The maximum number of lines of results to show in a table on an object page.

alleles_have_expression

If 1, allow editing of the expression of alleles and show the expression as a column in the genotype table on the genotype management page.

split_genotypes_by_organism

If 1, show the organism in a selector on the genotype management pages and only show genes and genotypes for the selected organism.

show_genotype_management_genes_list

If 0, hide the list of genes and actions on the left hand side of the genotype management page.

notes_on_single_genotypes_only

If 1, allow notes/comments on diploids and multi-locus genotypes.

allow_single_wildtype_allele

If 0, a wildtype allele is allowed only if there is another non-wildtype or non-wildtype expression allele from the same gene.

show_quick_deletion_buttons

If 1 (the default), show the "Deletion" buttons in the gene table on the genotype management page.

diploid_mode

If 1, allow diploids to be created and used in genotypes.

allele_note_types

A list of allele note type name and display names. The display_name is shown in the note editing dialog. Example configuration:

allele_note_types: - name: phenotype display_name: Phenotype - name: interaction_comment display_name: Interaction coment

db_initial_data

Data needed to initialise a Canto instance.

class_info

Descriptions of table in the database used by the interface. This information is used for rendering the view and object pages.

reports

A list of report names to show on the front page.

export

Configuration for exporting.

load

Configuration for loading data.

track_db_template_file

The template database to use when creating a new Canto instance.

curs_db_template_file

The template database to use when creating a new curation session.

ontology_index_dir

The name of the directory used for the ontology Lucene index. This index is used to do autocompeletion in the interface.

external_sources

URLs of external services.

implementation_classes

Names of classes used to implement database query and storage. This allows the implementations to be swapped from the defaults.

gene_adaptor

Used to look up gene identifier, name, synonyms and products. The default is to use the internal Canto database ("track").

evidence_types

Short name (codes) and long names of evidence types. Any evidence type configured with the option "with_gene" set to true will cause the interface to ask for a gene for later storage in the "with/from" column of a GAF file.

Evidence codes must also be configured for in available_annotation_type_list. See evidence_codes in the Configuring annotation types section.

Example:

evidence_types:
  IMP:
    name: Inferred from Mutant Phenotype
  IDA:
    name: Inferred from Direct Assay
  IGI:
    name: Inferred from Genetic Interaction
    with_gene: 1
  Microscopy: ~

In this case "Microscopy" is treated as an evidence code and an evience description.

namespace_term_evidence_codes

A map of defaults for the term to evidence lists configuration. If term_evidence_codes isn't set in an annotation type definition this map will be consulted to find defaults for the namespace of the annotation type. See term_evidence_codes for details.

This example will set term_evidence_codes for all annotation types that are configured with the namespace "fission_yeast_phenotype":

namespace_term_evidence_codes:
  fission_yeast_phenotype:
    - constraint: "is_a(FYPO:0001985)"
      evidence_codes:
       - Cell growth assay
       - Chromatin immunoprecipitation experiment

available_annotation_type_list

List of possible annotation type and their configuration details.

See Configuring annotation types for details.

enabled_annotation_type_list

The names of the types that are enabled in this Canto instance. If not set all annotation types from available_annotation_type_list will be enabled.

test_config_file

The path to the extra configuration file needed while testing.

help_text

The keys under help_text identify a part of Canto's user interface for which the help text should apply. Under the key is one or both of inline or url. The help text and link are rendered as a help icon with a tooltip that appears on mouse-over. If a url is given, clicking the help icon will send the user to the page specified by the url.

contact_email

This email address is shown anytime a contact address is needed. See the contact.mhtml template.

email_signature

A custom signature line that is displayed at the end of the generic email templates. If not set, the signature will default to "The [database_name] team", where the database name is the value of the database_name setting.

privacy_policy_url

The URL to the privacy policy for the database that runs the Canto instance. If configured, the following message will be shown on the Curator details page where the curator is asked to enter their personal details:

Your name, email and ORCID ID will be used under the terms of the [database_name] privacy policy. By continuing past this page, you agree to the terms of the policy.

The text 'privacy policy' will be linked to the privacy policy URL.

external_links

Each external link configuration has three possible parameters:

• name - The text to use as the title of the link.
• icon - An image from the root/static/images/logos to use as the <img> in the link.
• url - The URL to the appropriate page on an external site. This string can contain text to substitute in the form @@key@@. On the gene page, the key can be primary_identifier (the gene systematic ID) or primary_name (the gene name). On the publication page the key should be uniquename which will substitute the PubMed ID into the URL.
• sysid_link - This external link will be used to hyperlink the primary identifier on the gene page as well as being shown in the external links section on the right of the page.

There are two possible types of external link on the gene pages:

• generic - These links will be shown on all gene pages.
• organism - These links are specific to the given organism. The keys under the organism section will be the full organism name (ie "Genus species").

The external links are implemented in the linkouts.html template.

Configuring annotation types

The possible annotation type are set with the available_annotation_type_list configuration. It should be set to a YAML list of maps. An example is provided in the canto.yaml but can be overridden in the instance specific canto_deploy.yaml.

In canto_deploy.yaml you can add an optional enabled_annotation_type_list. If set only the listed types will be enabled.

For example this would enable just two annotation types:

enabled_annotation_type_list:
  - molecular_function
  - cellular_component

enabled_annotation_type_list exists so that one available_annotation_type_list configuration in canto.yaml can be used in several Canto instances. It is optional. If not set, all available types are enabled.

Each annotation type in available_annotation_type_list has the following settings:

name

A short internal name or key for this type. (Required)

category

This setting selects which code to use for the interface for this type. There are currently two possibilities: ontology and interaction. (Required)

display_name

The long name to display in the user for this type. eg. for molecular_function PomBase displays GO molecular function. (Required)

short_display_name

A shorter name for this type. eg molecular function (Optional)

very_short_display_name

A very short name for this type. eg F for GO molecular function. (Optional)

abbreviation

Currently used only for types from GO (molecular_function, cellular_component, and biological_process), the abbreviation is used from exporting annotation to GAF files. Required only if GO ontologies are enabled.

can_have_conditions

If 1, this annotation type can have conditions as well as evidence. Mostly useful for phenotypes.

single_allele_only

If true, only display/allow genotypes containing a single allele for this annotation type. If set to "ignore_accessory", accessory alleles will be ignored when counting alleles in a genotype.

single_locus_only

For interactions, restrict the genotype B list to contain only genotypes for the same locus selected in the genotype A list.

interaction_term_required

For interactions, selecting a term is required if and only if this option is set to 1.

second_feature_organism_selector

For interactions, if set the annotation dialog will have a independent organism selector for the second feature of the interaction. Defaults to 0 (false).

evidence_codes

Possible evidence codes (or interaction types) for this annotation type. Each evidence code must appear in the evidence_types list. (Required)

admin_evidence_codes

Evidence codes that are only available for logged in admin user. This list is added to the evidence_codes list.

term_evidence_codes

Used to restrict the visible evidence codes based on a currently selected term (if any). The constraint values are used for matching the term. If the current term is a descendent of the term before the "-" in the key and not a descendent of the terms to the right of the "-" the given evidence codes are shown to the user. The excluded terms to the right of the "-" are optional.

The first matching configuration is used. If the current term doesn't match any of the keys in term_evidence_codes, the default evidence codes from evidence_codes are displayed.

example:

term_evidence_codes:
  - constraint: "is_a(FYPO:0001985)"
    evidence_codes:
    - Cell growth assay
    - Chromatin immunoprecipitation experiment
    - Chromatography evidence

or:

term_evidence_codes:
  - constraint: "is_a(FYPO:0001985)-is_a(FYPO:0000045)&is_a(FYPO:0000150)"
    evidence_codes:
    - Cell growth assay
    - Chromatography evidence

See also namespace_term_evidence_codes.

term_suggestions_annotation_type

If set to the name of an ontology annotation type, use terms from the annotations of that type as suggestions. For example, once a genotype is selected in the interaction dialog, any phenotypes annotated for that genotype will be shown to the user. If unset, no suggestions will be shown.

hide_extension_relations

A list of extension relation names to hide. For hidden relations we just show the extension value (extension range). (default: empty list)

broad_term_suggestions

A few comma separated high level or broad term names for use in help text. eg. for molecular function: "transporter, transferase activity" (Required only for ontology annotation types)

specific_term_examples:

Comma separated specific term name examples. eg. "adenylate cyclase activity or biotin transporter activity" (Required only for ontology annotation types)

help_text

Short help to be soon initially to users when they begin an annotation of this type. (Required)

more_help_text

A longer help text shown when the user clicks "more..." under the help_text. (Optional)

detailed_help_path

A path in the documentation directory (root/docs/) to link to for detailed help about this type of annotation.

annotation_comment_help_text

Help text to show once the user finishes annotating a paper, after "Submit to curators". (Optional)

host_organism_taxonids

A list of taxon IDs for the organisms that are treated as host species in pathogen-host mode. All the list taxon IDs need to match organisms in the organism table. The organisms are loaded using using canto_add.pl --organism. See the Canto setup documentation for details.

If this list has any elements, Canto will start in pathogen-host mode and the internal config setting pathogen_host_mode will be automatically set to 1. Also host_organisms will be set to a list the Organism objects. multi_organism_mode is also set to 1/true.

Note that when pathogen_host_mode is enabled, every organism that is in the organism table but not in host_organism_taxonids will be assumed to be a pathogen organism.

allele_type_list

This list contains the configuration for each allele type

name

The allele type name to show in the display. This is also the key when looking up configuration details so it must be unique.

export_type

The type name to use when writing the export file. See Exporting data from Canto for more.

show_description

If true, show the description input box in the allele edit dialog.

description_required

If true, an allele of this type can't be created without a description.

allele_name_required

If true, alleles of this types must have a name.

allow_expression_change

If true, show the checkboxes for changing the allele expression.

expression_required

If true, an allele of this type can't be created without setting an expression level.

autopopulate_name

This template is used when an allele name can be automatically generated for a given type. For example, if the user selects "wild type" in the allele editing dialog, this template when set the name to something like "cdc2+". The string "@@gene_display_name@@" will be replaced with the current gene's name or systematic ID.

do_not_annotate

If true, don't show genotypes containing just this allele in the genotype selectors. And ignore alleles of this type when deciding if a genotype is single or multi allele.

evidence_code_groups

For the genotype_interaction annotation type we need to configure which interaction types are permitted for a given double mutant phenotype and for the alleles of the double mutant.

There are three sub-attributes:

double_mutant_population_phenotype

Settings and constraints for the case where the double mutant has a population phenotype.

parent_constraint

The term constraint used of populations phenotype terms.

evidence_codes

Interaction types to allow when the double mutant is a viable poputatation phenotype.

inviable_parent_constraint

The term constraint for inviable population phenotypes.

inviable_only_evidence_codes

Interaction types to allow when the double mutant is inviable.

not_population_evidence_codes

Types allowed when the double mutant doesn't have a poputation phenotype.

both_alleles_deletions

Interaction types allowed when both alleles are deletions.

one_allele_overexpressed

Interaction types allowed when at least one allele is overexpressed.