Class: Cell

One side of the tested comparison: a level of a model term, and which way that level entered. The unit a contrast is built from.

A cell points at a ModelTerm. The term's FactorLevels record which cohort, condition, occasion, or arm carries each level. Joining through the model recovers the contrast's entities, including its conditions and cohorts. The term is the comparison axis: cells sharing a term compare within a factor, cells on different terms compare across them.

A cell over a term that declares no levels -- a slope, or a product column -- leaves level unset and carries its direction alone.

URI: neurostudy:Cell

 classDiagram
    class Cell
    click Cell href "../Cell/"
      Cell : direction





        Cell --> "1" Direction : direction
        click Direction href "../Direction/"



      Cell : label

      Cell : level

      Cell : term





        Cell --> "1" ModelTerm : term
        click ModelTerm href "../ModelTerm/"

Slots

Name	Cardinality and Range	Description	Inheritance
term	1 ModelTerm	The model term this cell belongs to, by identifier	direct
level	0..1 String	The level represented by this cell, matching a `FactorLevel	direct
direction	1 Direction	How this cell entered the contrast: the weight's sign for a categorical level...	direct
label	0..1 String	The source's wording for this level in this comparison, when it differs from ...	direct

Unique Keys

level_per_term_per_effect

Unique key slots: term, level

Usages

used by	used in	type	used
Effect	cells	range	Cell

Identifier and Mapping Information

Schema Source

• from schema: https://neurostuff.org/schemas/neuroimaging-study-storage

Mappings

Mapping Type	Mapped Value
self	neurostudy:Cell
native	neurostudy:Cell

LinkML Source

Direct

name: Cell
description: 'One side of the tested comparison: a level of a model term, and which
  way that level entered. The unit a contrast is built from.

  A cell points at a `ModelTerm`. The term''s `FactorLevel`s record which cohort,
  condition, occasion, or arm carries each level. Joining through the model recovers
  the contrast''s entities, including its conditions and cohorts. The `term` is the
  comparison axis: cells sharing a term compare within a factor, cells on different
  terms compare across them.

  A cell over a term that declares no levels -- a slope, or a product column -- leaves
  `level` unset and carries its `direction` alone.'
from_schema: https://neurostuff.org/schemas/neuroimaging-study-storage
rank: 1000
attributes:
  term:
    name: term
    description: The model term this cell belongs to, by identifier. It must be one
      of the terms of the ModelEstimation the containing Analysis references, or of
      a stage that model reaches through `inputs_from` -- a group contrast of a first-level
      condition, or of a seed's time series, is a cell on that stage's term rather
      than on a copy of it.
    in_subset:
    - mvp
    - model_extracted
    from_schema: https://neurostuff.org/schemas/neuroimaging-study-storage/analysis
    rank: 1000
    domain_of:
    - Cell
    range: ModelTerm
    required: true
    inlined: false
  level:
    name: level
    description: 'The level represented by this cell, matching a `FactorLevel.level`
      on the referenced ModelTerm. The model states the level vocabulary; this field
      only identifies the level placed on this side of the comparison.

      Required exactly when the referenced term declares levels. A slope declares
      none, and its cell names none. A product column declares none either, but a
      signed cell on one crossing a factor with a continuous term may name a level
      of the factor: the cohort or condition whose slope has the cell''s sign, relative
      to the factor''s other levels. "Maltreatment was negatively associated with
      rACC activation in the PTSD group, but not in controls" is `direction: negative`
      with `level: PTSD group`. Without the level the sign has no reference, since
      which level the difference is taken from is a coding choice.'
    in_subset:
    - mvp
    - model_extracted
    from_schema: https://neurostuff.org/schemas/neuroimaging-study-storage/analysis
    rank: 1000
    domain_of:
    - Cell
    - FactorLevel
    range: string
  direction:
    name: direction
    description: 'How this cell entered the contrast: the weight''s sign for a categorical
      level, the fitted coefficient''s sign for a continuous term, or the interaction
      coefficient''s sign for a product column -- read as the named level''s slope
      relative to the others when the cell names one (`level`).

      Signs, never numeric weights. For an ordered comparison mark the extremes `positive`
      and `negative`, leave the levels in between out of `cells` -- absence is the
      zero weight -- and keep the ordering on `FactorLevel.order`.'
    in_subset:
    - mvp
    - model_extracted
    from_schema: https://neurostuff.org/schemas/neuroimaging-study-storage/analysis
    rank: 1000
    domain_of:
    - Cell
    range: Direction
    required: true
  label:
    name: label
    description: The source's wording for this level in this comparison, when it differs
      from the level's own label on the model.
    in_subset:
    - mvp
    - model_extracted
    from_schema: https://neurostuff.org/schemas/neuroimaging-study-storage/analysis
    rank: 1000
    domain_of:
    - Cell
    - DecodingClass
    range: string
unique_keys:
  level_per_term_per_effect:
    unique_key_name: level_per_term_per_effect
    unique_key_slots:
    - term
    - level

Induced

name: Cell
description: 'One side of the tested comparison: a level of a model term, and which
  way that level entered. The unit a contrast is built from.

  A cell points at a `ModelTerm`. The term''s `FactorLevel`s record which cohort,
  condition, occasion, or arm carries each level. Joining through the model recovers
  the contrast''s entities, including its conditions and cohorts. The `term` is the
  comparison axis: cells sharing a term compare within a factor, cells on different
  terms compare across them.

  A cell over a term that declares no levels -- a slope, or a product column -- leaves
  `level` unset and carries its `direction` alone.'
from_schema: https://neurostuff.org/schemas/neuroimaging-study-storage
rank: 1000
attributes:
  term:
    name: term
    description: The model term this cell belongs to, by identifier. It must be one
      of the terms of the ModelEstimation the containing Analysis references, or of
      a stage that model reaches through `inputs_from` -- a group contrast of a first-level
      condition, or of a seed's time series, is a cell on that stage's term rather
      than on a copy of it.
    in_subset:
    - mvp
    - model_extracted
    from_schema: https://neurostuff.org/schemas/neuroimaging-study-storage/analysis
    rank: 1000
    owner: Cell
    domain_of:
    - Cell
    range: ModelTerm
    required: true
    inlined: false
  level:
    name: level
    description: 'The level represented by this cell, matching a `FactorLevel.level`
      on the referenced ModelTerm. The model states the level vocabulary; this field
      only identifies the level placed on this side of the comparison.

      Required exactly when the referenced term declares levels. A slope declares
      none, and its cell names none. A product column declares none either, but a
      signed cell on one crossing a factor with a continuous term may name a level
      of the factor: the cohort or condition whose slope has the cell''s sign, relative
      to the factor''s other levels. "Maltreatment was negatively associated with
      rACC activation in the PTSD group, but not in controls" is `direction: negative`
      with `level: PTSD group`. Without the level the sign has no reference, since
      which level the difference is taken from is a coding choice.'
    in_subset:
    - mvp
    - model_extracted
    from_schema: https://neurostuff.org/schemas/neuroimaging-study-storage/analysis
    rank: 1000
    owner: Cell
    domain_of:
    - Cell
    - FactorLevel
    range: string
  direction:
    name: direction
    description: 'How this cell entered the contrast: the weight''s sign for a categorical
      level, the fitted coefficient''s sign for a continuous term, or the interaction
      coefficient''s sign for a product column -- read as the named level''s slope
      relative to the others when the cell names one (`level`).

      Signs, never numeric weights. For an ordered comparison mark the extremes `positive`
      and `negative`, leave the levels in between out of `cells` -- absence is the
      zero weight -- and keep the ordering on `FactorLevel.order`.'
    in_subset:
    - mvp
    - model_extracted
    from_schema: https://neurostuff.org/schemas/neuroimaging-study-storage/analysis
    rank: 1000
    owner: Cell
    domain_of:
    - Cell
    range: Direction
    required: true
  label:
    name: label
    description: The source's wording for this level in this comparison, when it differs
      from the level's own label on the model.
    in_subset:
    - mvp
    - model_extracted
    from_schema: https://neurostuff.org/schemas/neuroimaging-study-storage/analysis
    rank: 1000
    owner: Cell
    domain_of:
    - Cell
    - DecodingClass
    range: string
unique_keys:
  level_per_term_per_effect:
    unique_key_name: level_per_term_per_effect
    unique_key_slots:
    - term
    - level