NAME

Bio::EnsEMBL::Registry

SYNOPSIS

use Bio::EnsEMBL::Registry;

my $registry = 'Bio::EnsEMBL::Registry';

$registry->load_all("configuration_file");

$gene_adaptor = $registry->get_adaptor( 'Human', 'Core', 'Gene' );

DESCRIPTION

All Adaptors are stored/registered using this module. This module should then be used to get the adaptors needed.

The registry can be loaded from a configuration file using the load_all method.

If a filename is passed to load_all then this is used. Else if the environment variable ENSEMBL_REGISTRY is set to the name on an existing configuration file, then this is used. Else if the file .ensembl_init in your home directory exist, it is used.

For the Web server ENSEMBL_REGISTRY should be set in SiteDefs.pm. This will then be passed on to load_all.

The registry can also be loaded via the method load_registry_from_db which given a database host will load the latest versions of the Ensembl databases from it.

The four types of registries are for db adaptors, dba adaptors, dna adaptors and the standard type.

db

These are registries for backwards compatibility and enable the subroutines to add other adaptors to connections.

e.g. get_all_db_adaptors, get_db_adaptor, add_db_adaptor, remove_db_adaptor are the old DBAdaptor subroutines which are now redirected to the Registry.

So if before we had

my $sfa = $self->adaptor()->db()->get_db_adaptor('blast');

We now want to change this to

my $sfa =
  Bio::EnsEMBL::Registry->get_adaptor( "human", "core", "blast" );

DBA

These are the stores for the DBAdaptors

The Registry will create all the DBConnections needed now if you set up the configuration correctly. So instead of the old commands like

my $db           = Bio::EnsEMBL::DBSQL::DBAdaptor->new(...);
my $exon_adaptor = $db->get_ExonAdaptor;

we should now have just

my $exon_adaptor =
  Bio::EnsEMBL::Registry->get_adaptor( "human", "core", "exon" );

DNA

This is an internal Registry and allows the configuration of a dnadb. An example here is to set the est database to get its dna data from the core database.

## set the est db to use the core for getting dna data.
# Bio::EnsEMBL::Utils::ConfigRegistry->dnadb_add( "Homo Sapiens",
#   "core", "Homo Sapiens", "est" );

adaptors

This is the registry for all the general types of adaptors like GeneAdaptor, ExonAdaptor, Slice Adaptor etc.

These are accessed by the get_adaptor subroutine i.e.

my $exon_adaptor =
  Bio::EnsEMBL::Registry->get_adaptor( "human", "core", "exon" );

METHODS

load_all

Will load the registry with the configuration file which is
obtained from the first in the following and in that order.

 1) If an argument is passed to this method, this is used as the
    name of the configuration file to read.

 2) If the environment variable ENSEMBL_REGISTRY is set, this is
    used as the name of the configuration file to read.

 3) If the file .ensembl_init exist in the home directory, it is
    used as the configuration file.

 Arg [1]    : (optional) string
              Name of file to load the registry from.

 Arg [2]    : (optional) integer
              If not 0, will print out all information.

 Arg [3]    : (optional) integer
              If not 0, the database connection will not be
              cleared, if 0 or if not set the database connections
              will be cleared (this is the default).

 Arg [4]:     (optional) boolean
              This option will turn off caching for slice features,
              so, every time a set of features is retrieved,
              they will come from the database instead of the
              cache.  This option is only recommended for advanced
              users, specially if you need to store and retrieve
              features.  It might reduce performance when querying
              the database if not used properly.  If in doubt, do
              not use it or ask in the developer mailing list.

 Arg [5]:     (optional) boolean
              This option will make load_all() throw if the configuration file
              is missing and cannot be guessed from the environment

 Example    : Bio::EnsEMBL::Registry->load_all();
 Returntype : Int count of the DBAdaptor instances which can be found in the 
              registry due to this method being called. Will never be negative
 Exceptions : Throws if $throw_if_missing is set and ($config_file is missing
              and cannot be guessed from the environment
 Status     : Stable

clear

Will clear the registry and disconnect from all databases.

 Example    : Bio::EnsEMBL::Registry->clear();
 Returntype : none
 Exceptions : none
 Status     : Stable

add_db

Arg [1]    : db (DBAdaptor) to add adaptor to.
Arg [2]    : name of the name to add the adaptor to in the registry.
Arg [3]    : The adaptor to be added to the registry.
Example    : Bio::EnsEMBL::Registry->add_db($db, "lite", $dba);
Returntype : none
Exceptions : none
Status     : At Risk.
           : This is here for backwards compatibility only and may
           : be removed eventually.  Solution is to make sure the
           : db and the adaptor have the same species and the call
           : is then no longer needed.

remove_db

Arg [1]    : db (DBAdaptor) to remove adaptor from.
Arg [2]    : name to remove the adaptor from in the registry.
Example    : my $db = Bio::EnsEMBL::Registry->remove_db($db, "lite");
Returntype : adaptor
Exceptions : none
Status     : At Risk.
           : This is here for backwards compatibility only and may
           : be removed eventually.  Solution is to make sure the
           : db and the adaptor have the same species and the call
           : is then no longer needed.

get_db

Arg [1]    : db (DBAdaptor) to get adaptor from.
Arg [2]    : name to get the adaptor for in the registry.
Example    : my $db = Bio::EnsEMBL::Registry->get_db("Human", "core", "lite");
Returntype : adaptor
Exceptions : See get_DBAdaptor()
Status     : At Risk.
           : This is here for backwards compatibility only and may
           : be removed eventually.  Solution is to make sure the
           : db and the adaptor have the same species then call
           : get_DBAdaptor instead.

get_all_db_adaptors

Arg [1]    : db (DBAdaptor) to get all the adaptors from.
Example    : my $db = Bio::EnsEMBL::Registry->get_all_db_adaptors($db);
Returntype : adaptor
Exceptions : none
Status     : At Risk.
           : This is here for backwards compatibility only and
           : may be removed eventually.  Solution is to make
           : sure the dbs all have the same species then call
           : get_all_DBAdaptors(-species => "human");

add_DBAdaptor

Arg [1]    : name of the species to add the adaptor to in the registry.
Arg [2]    : name of the group to add the adaptor to in the registry.
Arg [3]    : DBAdaptor to be added to the registry.
Example    : Bio::EnsEMBL::Registry->add_DBAdaptor("Human", "core", $dba);
Returntype : none
Exceptions : none
caller     : internal
Status     : Stable

get_DBAdaptor

Arg [1]    : name of the species to get the adaptor for in the registry.
Arg [2]    : name of the group to get the adaptor for in the registry.
Arg [3]    : if set will not give warnings when looking for alias.
Example    : $dba = Bio::EnsEMBL::Registry->get_DBAdaptor("Human", "core");
Returntype : DBAdaptor
Exceptions : If $species is not defined and if no valid internal name 
             could be found for $species. If thrown check your API and DB
             version 
Status     : Stable

get_all_DBAdaptors

Arg [SPECIES]: (optional) string 
                species name to get adaptors for
Arg [GROUP]  : (optional) string 
                group name to get adaptors for
Example      : 
              @dba =
                @{ Bio::EnsEMBL::Registry->get_all_DBAdaptors() };

              @human_dbas =
                @{ Bio::EnsEMBL::Registry->get_all_DBAdaptors(
                  -species => 'human'
                ) };

Returntype   : list of DBAdaptors
Exceptions   : none
Status       : Stable

get_all_DBAdaptors_by_connection

Arg [1]    : DBConnection used to find DBAdaptors
Returntype : reference to list of DBAdaptors
Exceptions : none
Example    : @dba = @{ Bio::EnsEMBL::Registry
                ->get_all_DBAdaptors_by_connection($dbc) };
Status     : Stable

get_all_DBAdaptors_by_dbname

Arg [1]    : string, name of database
Returntype : reference to list of DBAdaptors
Exceptions : none
Example    : @dba = @{ Bio::EnsEMBL::Registry
                ->get_all_DBAdaptors_by_dbname($dbname) };
Status     : Stable

remove_DBAdaptor

Arg [1]    : name of the species to get the adaptor for in the registry.
Arg [2]    : name of the group to get the adaptor for in the registry.
Example    : $dba = Bio::EnsEMBL::Registry->remove_DBAdaptor("Human", "core");
Returntype : none
Exceptions : none
Status     : At risk

reset_DBAdaptor

Arg [1]:     string - species e.g. homo_sapiens
Arg [2]:     string - DB group e.g. core
Arg [3]:     string - new dbname
Args [4-7]:  string - optional DB parameters, defaults to current db params if omitted
Arg [8]:     hashref - Hash ref of additional parameters e.g. eFG dnadb params for auto selecting dnadb
Usage :      $reg->reset_registry_db( 'homo_sapiens', 'core',
                'homo_sapiens_core_37_35j' );
Description: Resets a DB within the registry.
Exceptions:  Throws if mandatory params not supplied
             Throws if species name is not already seen by the registry
             Throws if no current DB for species/group available
Status :     At risk

add_DNAAdaptor

Arg [1]    : name of the species to add the adaptor to in the registry.
Arg [2]    : name of the group to add the adaptor to in the registry.
Arg [3]    : name of the species to get the dna from
Arg [4]    : name of the group to get the dna from
Example    : Bio::EnsEMBL::Registry->add_DNAAdaptor("Human", "estgene", "Human", "core");
Returntype : none
Exceptions : none
Status     : Stable

get_DNAAdaptor

Arg [1]    : name of the species to get the adaptor for in the registry.
Arg [2]    : name of the group to get the adaptor for in the registry.
Example    : $dnaAdap = Bio::EnsEMBL::Registry->get_DNAAdaptor("Human", "core");
Returntype : adaptor
Exceptions : none
Status     : Stable

add_adaptor

Arg [1]    : name of the species to add the adaptor to in the registry.
Arg [2]    : name of the group to add the adaptor to in the registry.
Arg [3]    : name of the type to add the adaptor to in the registry.
Arg [4]    : The DBAdaptor to be added to the registry.
Arg [5]    : (optional) Set to allow overwrites of existing adaptors.
Example    : Bio::EnsEMBL::Registry->add_adaptor("Human", "core", "Gene", $adap);
Returntype : none
Exceptions : none
Caller     : internal
Status     : Stable

add_switchable_adaptor

Arg [1]    : String name of the species to add its switchable adaptor into the registry
Arg [2]    : String name of the group to add its switchable adaptor into the registry
Arg [3]    : String name of the type to add its switchable adaptor into the registry
Arg [4]    : Reference switchable adaptor to insert
Arg [5]    : Boolean override any existing switchable adaptor
Example    : Bio::EnsEMBL::Registry->add_switchable_adaptor("Human", "core", "Gene", $my_other_source);
Returntype : None
Exceptions : Thrown if a valid internal name cannot be found for the given 
             name. If thrown check your API and DB version. Also thrown if
             no type, group or switchable adaptor instance was given

has_switchable_adaptor

Arg [1]    : String name of the species to add its switchable adaptor into the registry
Arg [2]    : String name of the group to add its switchable adaptor into the registry
Arg [3]    : String name of the type to add its switchable adaptor into the registry
Example    : Bio::EnsEMBL::Registry->has_switchable_adaptor("Human", "core", "Gene");
Returntype : Boolean indicating if a switchable adaptor is available for your submitted combination
Exceptions : Thrown if a valid internal name cannot be found for the given 
             name. If thrown check your API and DB version. Also thrown if
             no type, group or switchable adaptor instance was given

remove_switchable_adaptor

Arg [1]    : name of the species to remove its switchable adaptor from the registry
Arg [2]    : name of the group to remove its switchable adaptor from the registry
Arg [3]    : name of the type to remove its switchable adaptor from the registry
Example    : $adap = Bio::EnsEMBL::Registry->remove_switchable_adaptor("Human", "core", "Gene");
Returntype : The removed adaptor if one was removed. Otherwise undef
Exceptions : Thrown if a valid internal name cannot be found for the given 
             name. If thrown check your API and DB version. Also thrown if
             no type or group was given

get_adaptor

Arg [1]     : name of the species to add the adaptor to in the registry.
Arg [2]     : name of the group to add the adaptor to in the registry.
Arg [3]     : name of the type to add the adaptor to in the registry.
Example     : $adap = Bio::EnsEMBL::Registry->get_adaptor("Human", "core", "Gene");
Description : Finds and returns the specified adaptor. This method will also check
              if the species, group and adaptor combination satisfy a DNADB condition
              (and will return that DNADB's implementation). Also we check for 
              any available switchable adaptors and will return that if available.
Returntype  : adaptor
Exceptions  : Thrown if a valid internal name cannot be found for the given 
              name. If thrown check your API and DB version. Also thrown if
              no type or group was given
Status      : Stable

get_all_adaptors

Arg [SPECIES] : (optional) string 
                species name to get adaptors for
Arg [GROUP] : (optional) string 
                group name to get adaptors for
Arg [TYPE] : (optional) string 
                type to get adaptors for
Example    : @adaps = @{Bio::EnsEMBL::Registry->get_all_adaptors()};
Returntype : ref to list of adaptors
Exceptions : none
Status     : Stable

add_alias

Arg [1]    : name of the species to add alias for
Arg [2]    : name of the alias
Example    : Bio::EnsEMBL::Registry->add_alias("Homo Sapiens","Human");
Description: add alternative name for the species.
Returntype : none
Exceptions : none
Status     : Stable

remove_alias

Arg [1]    : name of the species to remove alias for
Arg [2]    : name of the alias
Example    : Bio::EnsEMBL::Registry->remove_alias("Homo Sapiens","Human");
Description: remove alternative name for the species.
Returntype : none
Exceptions : none
Status     : Stable

get_alias

Arg [1]    : name of the possible alias to get species for
Example    : Bio::EnsEMBL::Registry->get_alias("Human");
Description: get proper species name.
Returntype : species name
Exceptions : none
Status     : Stable

get_all_aliases

Arg [1]    : Species name to retrieve aliases for
             (may be an alias as well).
Example    : Bio::EnsEMBL::Registry->get_all_aliases('Homo sapiens');
Description: Returns all known aliases for a given species (but not the
             species name/alias that was given).
Returntype : ArrayRef of all known aliases
Exceptions : none
Status     : Development

alias_exists

Arg [1]    : name of the possible alias to get species for
Example    : Bio::EnsEMBL::Registry->alias_exists("Human");
Description: does the species name exist.
Returntype : 1 if exists else 0
Exceptions : none
Status     : Stable

set_disconnect_when_inactive

Example    : Bio::EnsEMBL::Registry->set_disconnect_when_inactive();
Description: Set the flag to make sure that the database connection is dropped if
             not being used on each database.
Returntype : none
Exceptions : none
Status     : Stable

set_reconnect_when_lost

Example    : Bio::EnsEMBL::Registry->set_reconnect_when_lost();
Description: Set the flag to make sure that the database connection is not lost before it's used.
             This is useful for long running jobs (over 8hrs).
Returntype : none
Exceptions : none
Status     : Stable

disconnect_all

Example    : Bio::EnsEMBL::Registry->disconnect_all();
Description: disconnect from all the databases.
Returntype : none
Exceptions : none
Status     : Stable

get_DBAdaptor_count

Example     : Bio::EnsEMBL::Registry->get_DBAdaptor_count();
Description : Returns the count of database adaptors currently held by 
              the registry
Returntype  : Int count of database adaptors currently known
Exceptions  : None

change_access

Will change the username and password for a set of databases.
if host,user or database names are missing then these are not checked.
So for example if you do not specify a database then ALL databases on
the specified  host and port will be changed.

Arg [1]    : name of the host to change access on
Arg [2]    : port number to change access on
Arg [3]    : name of the user to change access on
Arg [4]    : name of the database to change access on
Arg [5]    : name of the new user
Arg [6]    : new password

Example    : Bio::EnsEMBL::Registry->get_alias("Human");
Description: change username and password on one or more databases
Returntype : none
Exceptions : none
Status     : Stable

load_registry_from_url

Arg [1] : string $url
Arg [2] : (optional) integer
          If not 0, will print out all information.
Arg [3] : (optional) integer
        This option will turn off caching for slice features, so,
        every time a set of features is retrieved, they will come
        from the database instead of the cache. This option is only
        recommended for advanced users, specially if you need to
        store and retrieve features. It might reduce performance when
        querying the database if not used properly. If in doubt, do
        not use it or ask in the developer mailing list.

Example : load_registry_from_url(
          'mysql://anonymous@ensembldb.ensembl.org:3306');

          load_registry_from_url(
          'mysql://anonymous@ensembldb.ensembl.org:3306/homo_sapiens_core_65_37?group=core&species=homo_sapiens'
          );

          load_registry_from_url(
          'mysql://anonymous@ensembldb.ensembl.org:3306/homo_sapiens_core_65_37?group=core'
          );


Description: Will load the correct versions of the ensembl
             databases for the software release it can find on
             a database instance into the registry. Also adds
             a set of standard aliases. The url format is:
             mysql://[[username][:password]@]hostname[:port].  You
             can also request a specific version for the databases
             by adding a slash and the version number but your
             script may crash as the API version won't match the
             DB version.

             You can also specify a database name which will cause the 
             loading of a single DBAdaptor instance. Parameters are
             mapped from a normal URL parameter set to their DBAdaptor
             equivalent. Group must be defined.

Returntype : Int count of the DBAdaptor instances which can be found in the 
             registry

Exceptions : Thrown if the given URL does not parse according to the above 
             scheme and if the specified database cannot be connected to 
             (see L<load_registry_from_db> for more information)
Status     : Stable

load_registry_from_db

Arg [HOST] : string
              The domain name of the database host to connect to.

Arg [USER] : string
              The name of the database user to connect with.

Arg [PASS] : (optional) string
              The password to be used to connect to the database.

Arg [PORT] : (optional) integer
              The port to use when connecting to the database.

Arg [VERBOSE]: (optional) boolean
              Whether to print database messages. This includes a listing
              of all available species & databases.

Arg [SPECIES]: (optional) string
              By default, all databases that are found on the
              server and that corresponds to the correct release
              are probed for aliases etc.  For some people,
              depending on where they are in the world, this might
              be a slow operation.  With the '-species' argument,
              one may reduce the startup time by restricting the
              set of databases that are probed to those of a
              particular species.

              Note that the latin name of the species is required,
              e.g., 'homo sapiens', 'gallus gallus', 'callithrix
              jacchus' etc.  It may be the whole species name,
              or only the first part of the name, e.g. 'homo',
              'gallus', or 'callithrix'.  This will be used in
              matching against the name of the databases.

Arg [DB_VERSION]: (optional) integer
              By default, only databases corresponding to the
              current API version are loaded.  This argument
              allows the script to use databases from another
              version although it might not work properly.  This
              argument should only be used for production or
              testing purposes and if you really know what you are
              doing.

Arg [WAIT_TIMEOUT]: (optional) integer
              Time in seconds for the wait timeout to happen.
              Time after which the connection is deleted if not
              used.  By default this is 28800 (8 hours), so set
              this to greater than this if your connection are
              getting deleted.  Only set this if you are having
              problems and know what you are doing.

 Arg [-NO_CACHE]: (optional) boolean
              This option will turn off caching for slice features,
              so, every time a set of features is retrieved, they
              will come from the database instead of the cache.  This
              option is only recommended for advanced users, specially
              if you need to store and retrieve features.  It might
              reduce performance when querying the database if not
              used properly.  If in doubt, do not use it or ask in the
              developer mailing list.

 Arg [SPECIES_SUFFIX]: (optional) string
              This option will append the string to the species name
              in the registry for all databases found on this server.

Example :

  $registry->load_registry_from_db(
    -host    => 'ensembldb.ensembl.org',
    -user    => 'anonymous',
    -verbose => '1'
  );

Description: Will load the correct versions of the Ensembl
             databases for the software release it can find on a
             database instance into the registry.  Also adds a set
             of standard aliases.

Returntype : Int count of the DBAdaptor instances which can be found in the 
             registry due to this method call.

Exceptions : Thrown if the given MySQL database cannot be connected to
             or there is any error whilst querying the database.
Status     : Stable

_group_to_adaptor_class

Arg [1]       : The group you wish to decode to an adaptor class
Example       : Bio::EnsEMBL::Registry->_group_to_adaptor_class('core');
Description   : Has an internal lookup of groups to their adaptor classes
Returntype    : String
Exceptions    : Thrown if the group is unknown
Status        : Stable

find_and_add_aliases

Arg [ADAPTOR] : (optional) Bio::EnsEMBL::DBSQL::DBAdaptor
                The adaptor to use to retrieve aliases from.

Arg [GROUP]   : (optional) string
                The group you want to find aliases for. If not
                given assumes all types.

Arg [HANDLE]  : (optional) DBI database handle
                A connected database handle to use instead of
                the database handles stored in the DBAdaptors.
                Bypasses the use of MetaContainer.

Arg [SPECIES_SUFFIX]: (optional) string
                This option will append the string to the species
                name in the registry for all databases.

Example       : Bio::EnsEMBL::Registry->find_and_add_aliases(
                  -ADAPTOR => $dba,
                  -GROUP   => 'core'
                );

Description   : Looks in the meta container for each database for
                an entry called "species.alias".  If any are found
                then the species adaptor is registered to that
                set of aliases.  This can work across any adaptor
                which has a MetaContainer.  If no MetaContainer
                can be returned from a given adaptor then no alias
                searching is performed.

Return type   : none
Exceptions    : Throws if an alias is found in more than one species.
Status        : Stable

load_registry_from_multiple_dbs

Arg [1]   : Array of hashes, each hash being a set of arguments to
            load_registry_from_db() (see above).

Example   :

  $registry->load_registry_from_multiple_dbs( {
      '-host'    => 'ensembldb.ensembl.org',
      '-user'    => 'anonymous',
      '-verbose' => '1'
    },
    {
      '-host'     => 'server.example.com',
      '-user'     => 'anonymouse',
      '-password' => 'cheese',
      '-verbose'  => '1'
    } );

Description:  Will call load_registry_from_db() (see above)
              multiple times and merge the resulting registries
              into one, effectively allowing a user to connect to
              databases on multiple database servers from within
              one program.

              If a database is found on more than one server, the
              first found instance of that database will be used.

Returntype : Int count of the DBAdaptor instances which can be found in the 
             registry

set_default_track

Sets a flag to say that that this species/group are a default track and do not
need to be added as another web track.

Arg [1]    : name of the species to get the adaptors for in the registry.
Arg [2]    : name of the type to get the adaptors for in the registry.
Example    : $merged = Bio::EnsEMBL::Registry->set_default_track("Human","core");
Returntype : none
Exceptions : none
Status     : At Risk.

default_track

Check flag to see if this is a default track

Arg [1]    : name of the species to get the adaptors for in the registry.
Arg [2]    : name of the type to get the adaptors for in the registry.
Example    : $merged = Bio::EnsEMBL::Registry->set_default_track("Human","core");
Returntype : int 
Exceptions : none
Status     : At Risk.

add_new_tracks

Will add new gene tracks to the configuration of the WEB server if they are
not of the type default and the configuration already has genes in the display.

Arg [1]    : hash of the default configuration of the web page
Returntype : none
Exceptions : none
Called by  : UserConfig.pm
Status     : At Risk.

no_version_check

getter/setter for whether to run the version checking

Arg[0]     : (optional) int
Returntype : int or undef if not set
Exceptions : none
Status     : At Risk.

no_cache_warnings

Arg[0]      : boolean for turning the flag on and off
Description : Turns off any warnings about not using caching in all available 
              adaptors. 
Returntype  : boolean Current status
Exceptions  : None

version_check

run the database/API code version check for a DBAdaptor

Arg[0]     : DBAdaptor to check
Returntype : int 1 if okay, 0 if not the same 
Exceptions : none
Status     : At Risk.

get_all_species

Arg [1]    : String group type, such as core, or otherfeatures
Description: Method for getting all valid species names found in available
             databases. This excludes the ancestral sequence databases, and
             any species from a non-core database. Specifying a group allows
             the list to apply to non-core database types.
Example    : my @species_names = @{ $reg->get_all_species() };
Returntype : Listref of species names

get_species_and_object_type

Description:  Get the species name, object type (gene, transcript,
              translation, or exon etc.), and database type for a
              stable ID.

Arg [1]    :  String stable_id
              The stable ID to find species and object type for.

Arg [2]    :  String known_type (optional)
              The type of the stable ID, if it is known.

Arg [3]    :  String known_species (optional)
              The species, if known

Arg [4]    :  String known_db_type (optional)
              The database type, if known

Example    :  my $stable_id = 'ENST00000326632';

              my ( $species, $object_type, $db_type ) =
                $registry->get_species_and_object_type($stable_id);

              my $adaptor =
                $registry->get_adaptor( $species, $db_type,
                                        $object_type );

              my $object = $adaptor->fetch_by_stable_id($stable_id);

Return type:  Array consisting of the species name, object type,
              and database type.  The array may be empty if no
              match is found.

Exceptions :  none
Status     :  At Risk.