Class: Study

Overview

Note:

This is really quite convoluted, and couples together administrative organization alongside accessioning and data-access rules. It results in samples being tied to an EGAS/ERP far too early in their lifecycle, and as a result we often need to perform 'sample moves'. Although we do need to know if samples are open(ENA) or managed(EGA) at the point of accessioning.

A Study is a collection of various samples and the work done on them. They are perhaps slightly overloaded, and provide: - A means of grouping together samples for administrative purposes - A means of generating EGAS/ERP study accession numbers at the ENA/EGA - @see Accessionable::Study - These accession numbers are used at data release to group samples together for publication - For managed/EGA studies, also ties the data to an Accessionable::Dac and Accessionable::Policy - A means of generating the aforementioned Accessionable::Dac and Accessionable::Policy @note These should DEFINITELY be separate entities - A means of tying data to internal data-release timings - A means to apply internal data access policies to released sequencing data - A means to tie interested parties to the samples and the work done on them - A way of specifying common ways of filtering/processing generated data. eg. filter human sequence - The service with which a Sample will be accessioned (eg. EGA/ENA)

When a Sample enters Sequencescape it will usually be associated with a single Study, usually determined by the Study associated with the SampleManifest. This study will be recorded on the Aliquot in the stock Receptacle, and additionally a StudySample will record this association.

When work is requested an Order will be created, specifying a list of receptacles and the Study for which this work is being performed. This will set initial study id on request and in turn will be recorded on any downstream aliquots. Critically, it is the study specified on the Aliquot in the Lane which will influence processes like data release and data access.

Defined Under Namespace

Classes: Metadata

Constant Summary collapse

STOCK_PLATE_PURPOSES =

Constants

['Stock Plate', 'Stock RNA Plate'].freeze
YES =
'Yes'
NO =
'No'
YES_OR_NO =
[YES, NO].freeze
OTHER_TYPE =
'Other'
STUDY_SRA_HOLDS =
%w[Hold Public].freeze
DATA_RELEASE_STRATEGY_OPEN =
'open'
DATA_RELEASE_STRATEGY_MANAGED =
'managed'
DATA_RELEASE_STRATEGY_NOT_APPLICABLE =
'not applicable'
DATA_RELEASE_STRATEGIES =
[
  DATA_RELEASE_STRATEGY_OPEN,
  DATA_RELEASE_STRATEGY_MANAGED,
  DATA_RELEASE_STRATEGY_NOT_APPLICABLE
].freeze
DATA_RELEASE_TIMING_STANDARD =
'standard'
DATA_RELEASE_TIMING_NEVER =
'never'
DATA_RELEASE_TIMING_DELAYED =
'delayed'
DATA_RELEASE_TIMING_IMMEDIATE =
'immediate'
DATA_RELEASE_TIMING_PUBLICATION =
'delay until publication'
DATA_RELEASE_TIMINGS =
[
  DATA_RELEASE_TIMING_STANDARD,
  DATA_RELEASE_TIMING_IMMEDIATE,
  DATA_RELEASE_TIMING_DELAYED
].freeze
OLD_DATA_RELEASE_PREVENTION_REASONS =
['data validity', 'legal', 'replication of data subset'].freeze
DATA_RELEASE_PREVENTION_REASON_OTHER =
'Other (please specify)'
DATA_RELEASE_PREVENTION_REASONS =
[
  'Pilot or validation studies - DAC approval not required',
  'Collaborators will share data in a research repository - DAC approval not required',
  'Prevent harm (e.g sensitive studies or biosecurity) - DAC approval required',
  'Protecting IP - DAC approval required',
  DATA_RELEASE_PREVENTION_REASON_OTHER
].freeze
OLD_DATA_RELEASE_DELAY_FOR_OTHER =
'other'
DATA_RELEASE_DELAY_FOR_OTHER =
'Other (please specify below)'
OLD_DATA_RELEASE_DELAY_REASONS =
['other', 'phd study'].freeze
DATA_RELEASE_DELAY_REASONS_STANDARD =
[
  'PhD study',
  'Capacity building',
  'Intellectual property protection',
  'Additional time to make data FAIR',
  DATA_RELEASE_DELAY_FOR_OTHER
].freeze
DATA_RELEASE_DELAY_REASONS_ASSAY =
['assay of no other use'].freeze
DATA_RELEASE_DELAY_PERIODS =
['3 months', '6 months', '9 months', '12 months', '18 months'].freeze
EBI_LIBRARY_STRATEGY_OPTIONS =
EBI_LIBRARY_SOURCE_OPTIONS =
EBI_LIBRARY_SELECTION_OPTIONS =
REMAPPED_ATTRIBUTES =
{
  contaminated_human_dna: YES_OR_NO,
  remove_x_and_autosomes: YES_OR_NO,
  study_sra_hold: STUDY_SRA_HOLDS,
  contains_human_dna: YES_OR_NO,
  commercially_available: YES_OR_NO
}.transform_values { |v| v.index_by { |b| b.downcase } }

Constants included from Metadata

Metadata::SECTION_FIELDS

Constants included from StudyReport::StudyDetails

StudyReport::StudyDetails::BATCH_SIZE

Instance Attribute Summary collapse

Instance Method Summary collapse

Methods included from EventfulRecord

has_many_events, has_many_lab_events, has_one_event_with_family

Methods included from Metadata

has_metadata

Methods included from Attributable::Association::Target

default, extended, for_select_association

Methods included from SampleManifest::Associations

included

Methods included from ReferenceGenome::Associations

included

Methods included from SharedBehaviour::Named

included

Methods included from Commentable

#after_comment_addition

Methods included from DataRelease

#accession_required?, #do_not_enforce_accessioning, #for_array_express?, #valid_data_release_properties?

Methods included from Uuid::Uuidable

included, #unsaved_uuid!, #uuid

Methods included from Api::StudyIo::Extensions

included, #render_class

Methods included from StudyReport::StudyDetails

#each_stock_well_id_in_study_in_batches, #progress_report_header, #progress_report_on_all_assets

Methods inherited from ApplicationRecord

alias_association, convert_labware_to_receptacle_for, find_by_id_or_name, find_by_id_or_name!

Methods included from Squishify

extended

Instance Attribute Details

#approvalObject

Returns the value of attribute approval.



128
129
130
# File 'app/models/study.rb', line 128

def approval
  @approval
end

#run_countObject

Returns the value of attribute run_count.



128
129
130
# File 'app/models/study.rb', line 128

def run_count
  @run_count
end

#total_priceObject

Returns the value of attribute total_price.



128
129
130
# File 'app/models/study.rb', line 128

def total_price
  @total_price
end

Instance Method Details

#abbreviationObject



562
563
564
565
# File 'app/models/study.rb', line 562

def abbreviation
  abbreviation = .study_name_abbreviation
  abbreviation.presence || "#{id}STDY"
end

#accession_all_samplesObject



558
559
560
# File 'app/models/study.rb', line 558

def accession_all_samples
  samples.find_each(&:accession) if accession_number?
end

#accession_number?Boolean

Returns:

  • (Boolean)


554
555
556
# File 'app/models/study.rb', line 554

def accession_number?
  ebi_accession_number.present?
end

#accession_serviceObject



581
582
583
584
585
586
587
588
589
590
# File 'app/models/study.rb', line 581

def accession_service
  case data_release_strategy
  when 'open'
    EnaAccessionService.new
  when 'managed'
    EgaAccessionService.new
  else
    NoAccessionService.new(self)
  end
end

#approved?Boolean

Returns:

  • (Boolean)


571
572
573
574
# File 'app/models/study.rb', line 571

def approved?
  # TODO: remove
  true
end

#asset_progress(assets = nil) {|initial_requests.asset_statistics(wheres)| ... } ⇒ Object

Yields information on the state of all assets in a convenient fashion for displaying in a table.

Yields:

  • (initial_requests.asset_statistics(wheres))


492
493
494
495
496
# File 'app/models/study.rb', line 492

def asset_progress(assets = nil)
  wheres = {}
  wheres = { asset_id: assets.map(&:id) } if assets.present?
  yield(initial_requests.asset_statistics(wheres))
end

#completedObject



477
478
479
480
481
482
483
# File 'app/models/study.rb', line 477

def completed
  counts = requests.standard.group('state').count
  total = counts.values.sum
  failed = counts['failed'] || 0
  cancelled = counts['cancelled'] || 0
  (total - failed - cancelled) > 0 ? (counts.fetch('passed', 0) * 100) / (total - failed - cancelled) : 0
end

#dac_accession_numberObject



546
547
548
# File 'app/models/study.rb', line 546

def dac_accession_number
  .ega_dac_accession_number
end

#dac_refnameObject



515
516
517
# File 'app/models/study.rb', line 515

def dac_refname
  "DAC for study - #{name} - ##{id}"
end

#data_release_delay_options(assay_option: false) ⇒ Array<String>

Helper method for edit dropdowns to support backwards compatibility with old options.

Parameters:

  • assay_option (Boolean) (defaults to: false)
    • whether to include assay-specific options

Returns:

  • (Array<String>)

    the list of options for the data release delay reason dropdown



629
630
631
632
633
634
635
636
637
638
# File 'app/models/study.rb', line 629

def data_release_delay_options(assay_option: false)
  # If the current value is an old one, then we need to include it in the list of options
  additional_options = []
  if OLD_DATA_RELEASE_DELAY_REASONS.include? .data_release_delay_reason
    additional_options << .data_release_delay_reason
  end

  additional_options.concat(DATA_RELEASE_DELAY_REASONS_ASSAY) if assay_option
  DATA_RELEASE_DELAY_REASONS_STANDARD + additional_options
end

#data_release_prevention_optionsArray<String>

Helper method for edit dropdowns to support backwards compatibility with old options.

Returns:

  • (Array<String>)

    the list of options for the data release prevention reason dropdown



616
617
618
619
620
621
622
623
# File 'app/models/study.rb', line 616

def data_release_prevention_options
  additional_options = []
  if OLD_DATA_RELEASE_PREVENTION_REASONS.include? .data_release_prevention_reason
    additional_options << .data_release_prevention_reason
  end

  DATA_RELEASE_PREVENTION_REASONS + additional_options
end

#dehumanise_abbreviated_nameObject



567
568
569
# File 'app/models/study.rb', line 567

def dehumanise_abbreviated_name
  abbreviation.downcase.gsub(/ +/, '_')
end

#each_well_for_qc_report_in_batches(exclude_existing, product_criteria, plate_purposes = nil) ⇒ Object



442
443
444
445
446
447
448
449
450
451
452
453
454
# File 'app/models/study.rb', line 442

def each_well_for_qc_report_in_batches(exclude_existing, product_criteria, plate_purposes = nil)
  # @note We include aliquots here, despite the fact they are only needed if we have to set a poor-quality flag
  #       as in some cases failures are not as rare as you may imagine, and it can cause major performance issues.
  base_scope =
    Well
      .on_plate_purpose_included(PlatePurpose.where(name: plate_purposes || STOCK_PLATE_PURPOSES))
      .for_study_through_aliquot(self)
      .without_blank_samples
      .includes(:well_attribute, :aliquots, :map, samples: :sample_metadata)
      .readonly(true)
  scope = exclude_existing ? base_scope.without_report(product_criteria) : base_scope
  scope.find_in_batches { |wells| yield wells }
end

#ebi_accession_numberObject



542
543
544
# File 'app/models/study.rb', line 542

def ebi_accession_number
  .study_ebi_accession_number
end

#ethical_approval_required?Boolean

Returns:

  • (Boolean)


576
577
578
579
# File 'app/models/study.rb', line 576

def ethical_approval_required?
  .contains_human_dna == Study::YES && .contaminated_human_dna == Study::NO &&
    .commercially_available == Study::NO
end

#localeObject



538
539
540
# File 'app/models/study.rb', line 538

def locale
  funding_source
end

#mailing_list_of_managersObject



600
601
602
603
# File 'app/models/study.rb', line 600

def mailing_list_of_managers
  configured_managers = managers.pluck(:email).compact.uniq
  configured_managers.empty? ? configatron.fetch(:ssr_emails, User.all_administrators_emails) : configured_managers
end

#mark_activeObject



469
470
471
# File 'app/models/study.rb', line 469

def mark_active
  logger.warn "Study activation failed! #{errors.map(&:to_s)}" unless active?
end

#mark_deactiveObject



465
466
467
# File 'app/models/study.rb', line 465

def mark_deactive
  logger.warn "Study deactivation failed! #{errors.map(&:to_s)}" unless inactive?
end

#ownerObject

Returns the study owner (user) if exists or nil TODO - Should be “owners” and return all owners or empty array - done TODO - Look into this is the person that created it really the owner? If so, then an owner should be created when a study is created.



534
535
536
# File 'app/models/study.rb', line 534

def owner
  owners.first
end

#policy_accession_numberObject



550
551
552
# File 'app/models/study.rb', line 550

def policy_accession_number
  .ega_policy_accession_number
end

#rebroadcastObject



609
610
611
# File 'app/models/study.rb', line 609

def rebroadcast
  broadcast
end

#request_progress {|@stats_cache ||= initial_requests.progress_statistics| ... } ⇒ Object

Yields information on the state of all request types in a convenient fashion for displaying in a table. Used initial requests, which won't capture cross study sequencing requests.

Yields:

  • (@stats_cache ||= initial_requests.progress_statistics)


487
488
489
# File 'app/models/study.rb', line 487

def request_progress
  yield(@stats_cache ||= initial_requests.progress_statistics) if block_given?
end

#sample_progress(samples = nil) ⇒ Object

Yields information on the state of all samples in a convenient fashion for displaying in a table.



499
500
501
502
503
504
505
506
507
508
509
# File 'app/models/study.rb', line 499

def sample_progress(samples = nil)
  if samples.blank?
    requests.sample_statistics_new
  else
    # Rubocop suggests this changes as it allows MySQL to perform a single query, which is usually better
    # however in this case we've actually already loaded the samples. If we do try passing in the
    # samples themselves, then things top working as intended. (Performance tanks in some places, and
    # we generate invalid SQL in others)
    yield(requests.where(aliquots: { sample_id: samples.pluck(:id) }).sample_statistics_new)
  end
end

#send_samples_to_service?Boolean

Returns:

  • (Boolean)


592
593
594
# File 'app/models/study.rb', line 592

def send_samples_to_service?
  accession_service.no_study_accession_needed || (!.never_release? && accession_number?)
end

#studyObject

Used by EventfulMailer



525
526
527
# File 'app/models/study.rb', line 525

def study
  self
end

#study_statusObject



511
512
513
# File 'app/models/study.rb', line 511

def study_status
  inactive? ? 'closed' : 'open'
end

#subject_typeObject



605
606
607
# File 'app/models/study.rb', line 605

def subject_type
  'study'
end

#text_commentsObject



473
474
475
# File 'app/models/study.rb', line 473

def text_comments
  comments.each_with_object([]) { |c, array| array << c.description if c.description.present? }.join(', ')
end

#unprocessed_submissions?Boolean

Returns:

  • (Boolean)


519
520
521
522
# File 'app/models/study.rb', line 519

def unprocessed_submissions?
  # TODO[mb14] optimize if needed
  study.orders.any? { |o| o.submission.nil? || o.submission.unprocessed? }
end

#validate_ena_required_fields!Object



596
597
598
# File 'app/models/study.rb', line 596

def validate_ena_required_fields!
  valid?(:accession) or raise ActiveRecord::RecordInvalid, self
end

#validate_ethically_approvedObject

Instance methods



429
430
431
432
433
434
435
436
437
438
439
440
# File 'app/models/study.rb', line 429

def validate_ethically_approved
  return true if valid_ethically_approved?

  message =
    if ethical_approval_required?
      'should be either true or false for this study.'
    else
      'should be not applicable (null) not false.'
    end
  errors.add(:ethically_approved, message)
  false
end

#warningsObject



456
457
458
459
460
461
462
463
# File 'app/models/study.rb', line 456

def warnings
  # These studies are now invalid, but the warning should remain until existing studies are fixed.
  if .managed? && .data_access_group.blank?
    # rubocop:todo Layout/LineLength
    'No user group specified for a managed study. Please specify a valid Unix user group to ensure study data is visible to the correct people.'
    # rubocop:enable Layout/LineLength
  end
end