Proposed Relations

From Program on Ontologies of Neural Structures

Contents 1 Proposed Relations 1.1 Relation 1 1.1.1 List of potential names (currently preferred name in italics) 1.1.2 Proposed formalization: 1.1.3 Examples: 1.2 Relation 2 1.2.1 List of potential names (currently preferred name in italics) 1.2.2 Proposed formalization 1.2.3 Examples: 1.3 Relation 3: 1.3.1 Potential names 1.3.2 Proposed formalization 1.3.3 Examples: 1.3.4 Notes 1.4 Relation 4 1.4.1 Potential names (currently preferred name in italics) 1.4.2 Proposed formalization 1.4.3 Examples: 1.4.4 Notes 1.5 Relation 5 1.5.1 List of potential names (currently preferred name in italics). 1.5.2 Proposed formalization: 1.5.3 Examples: 1.6 Relation 6 1.6.1 Potential names 1.6.2 Proposed formalization 1.6.3 Examples: 1.7 Relation 7 1.7.1 Potential names 1.7.2 Proposed formalization 1.7.3 Examples: 1.8 Relation 8 1.8.1 Potential names (currently preferred name in italics) 1.8.2 Proposed formalization 1.8.3 Examples: 1.8.4 Notes: 1.9 Relation 9 1.9.1 Proposed names: 1.9.2 Proposed formalization 1.10 Relation 10 1.10.1 Potential names: 1.10.2 Proposed formalization 1.10.3 Examples:

Proposed Relations

For an OWL version of these relations, please see http://purl.obolibrary.org/obo/ro.owl

Relation 1

Relation between a neuron and an entity it chemically synapses to:

List of potential names (currently preferred name in italics)

1. synapses_to
2. synapsed_to
3. is_synapsed_to
4. chemically_synapsed_to
5. chemically_synapses_to
6. is_chemically_synapsed_to
7. chemically_synapses_to

• inverse_of Relation 2

Proposed formalization:

'Every N1 is_synapsed_to some N2' expands to:

  Every N1 has_part some (
     'pre-synaptic membrane ; GO:0042734' that part_of some ( 
        'synapse ; GO:0045202' that has_part some (
           'post-synaptic membrane ; GO:0045211' that part_of some N2 ))))

domain: (neuron or part_of some neuron)

Examples:

1. Every 'olfactory receptor neuron' is_synapsed_to some 'antennal lobe projection neuron'
2. Every 'somatic motor neuron' is_synapsed_to some 'skeletal muscle'

Relation 2

Relation between an entity and a neuron that is chemically synapsed to it

List of potential names (currently preferred name in italics)

1. synapsed_by
2. is_synapsed_by
3. is_chemically_synapsed_by
4. chemically_synapsed_by

• inverse_of Relation 1

Proposed formalization

'Every N1 is_synapsed_by some N2' expands to:

  Every N1 has_part some (
     'post-synaptic membrane ; GO:0045211' that part_of some ( 
        'synapse ; GO:0045202' that has_part some (
           'pre-synaptic membrane ; GO:0042734' that part_of some N2))))

range: (neuron or part_of some neuron)

Examples:

1. Every interneuron is_synapsed_by some neuron

Relation 3:

Relation between a neuron and an anatomical structure (such as a brain region) in which it (chemically) synapses to targets.

Potential names

1. synapses_in
   • cons: not clear what should be the inverse relation name
2. has_pre-synaptic_terminal_in

Proposed formalization

'Every N synapses_in some R' expands_to:

  Every N has_part some (
     'pre-synaptic membrane ; GO:0042734' that part_of some (
        'synapse ; GO:0045202' that part_of some R )))

domain: (neuron or part_of some neuron)

Rule: If n synapses_in r1 and r1 part_of r2 then n synapses_in r2.

Rule expressed as property chain: synapses_in o part_of -> synapases_in

Examples:

See Relation 6. (This relation is a more general version of Relation 6).

Notes

This is a more general relation than Relation 6. Unlike it, this relation doesn't depend on the neuronal processes involved being clearly distinguishable as axons. This makes in useful for capturing connectivity in invertebrate central nervous systems in cases where is not clear that the axon/dendrite distinction makes sense (a single neuron projection is receives multiple synaptic input and synapses to other neurons).

Relation 4

Relation between a neuron and an anatomical structure (such as a brain region) in which other neurons (chemically) synapse to it.

Potential names (currently preferred name in italics)

1. synapsed_in
   • con: not sure what should be the inverse relation name
2. is_synapsed_in
   • con: not sure what should be the inverse relation name
3. has_post-synaptic_terminals_in

Proposed formalization

'Every N is_synapsed_in some R' Expands_to:

  Every N has_part some (
     ‘post-synaptic membrane ; GO:0045211’ that part_of some
        ‘synapse ; GO:0045202’ that part_of some R )))

domain: (neuron or part_of some neuron)

Rule: If n synapsed_in r1 and r1 part_of r2 then n synapsed_in r2.

Rule expressed as property chain: synapsed_in o part_of -> synapased_in

Examples:

See Relation 7. This relation is a more general version of Relation 7.

Notes

This is a more general relation than Relation 7. Unlike it, this relation doesn't depend on the synapse being made to a dendrite. can't be used. This makes it useful in invertebrate central nervous systems for cases where it is not clear that the axon/dendrite distinction makes sense (a single neuron projection receives multiple synaptic input and synapses to other neurons). It could also be used in vertebrate nervous systems in cases where a dendrite synpases directly to the cell body of its target.

Relation 5

General relation between a neuron and an anatomical structure (such as a brain region) in which it either synapses to targets or is synapsed to or both.

List of potential names (currently preferred name in italics).

1. innervates (note - potential clash with preferred name for Relation 9)
   • con: 'innervates' is used much more generally – and more commonly for tract or nerve to region.
   • pro: easy to name inverse – innervated_by.
2. synapse_in
   • con: Name too similar to others suggested here.
   • con: difficult to name inverse
   • pro: it's not innervates, which has other meanings.
3. has_synapse_in
   • inverse: has_synapse_of ?
4. has_synaptic_terminals_in

Proposed formalization:

'Every X has_synapse_in_some Y' expands_to:

  Every X has part some (
     'postsynaptic membrane' and 'part of' some (
        'synapse' and 'part of' some Y
        )
     ) 
or (
     'has part' some (
        'presynaptic membrane' and 'part of' some (
           'synapse' and 'part of' some Y)))

Property hierarchy:

  has_synapse_in 
     <- sub_property_of synapsed_in
        <- sub_property_of dendrite_synapsed_in 
     <- sub_property_of synapses_in
        <- sub_property_of axon_synapses_in

domain: (neuron or part_of some neuron)

Rule: If n synapsed_in r1 and r1 part_of r2 then n synapsed_in r2.

Rule expressed as property chain: synapsed_in o part_of -> synapased_in

Examples:

1.  Every Kenyon cell has_synapse_in some pedunculus

Relation 6

Relation between a neuron and a region in which it axon(s) forms chemical synapses to targets.

• Note, this is sub-relation of Relation 3.

Potential names

1. axon_innervates
   • con: 'innervates' is used much more generally – and more commonly for tract or nerve to region.
   • pro: easy to name inverse – innervated by axon of.
2. axon_synapses_in
   • con: difficult to name inverse
   • pro: It doesn't use innervates, which has other meanings.

Proposed formalization

'Every N axon_synapses_in some R' expands_to:

  Every N has_part some (
     'axon ; GO:0030424' that has_part some (
        'pre-synaptic membrane ; GO:0042734' that part_of some (
           'synapse ; GO:0045202' that part_of some R))))

domain: neuron

Rule: If n axon_synapses_in r1 and r1 part_of r2 then n axon_synapses_in r2.

Rule expressed as property chain: axon_synapses_in o part_of -> axon_synapases_in

Examples:

1. Every 'projection neuron DL1 adPN' axon_innervates some 'lateral horn'
2. Every 'adult olfactory receptor neuron' axon_innervates some 'adult antennal lobe glomerulus'

Relation 7

Relation between a neuron and a region in which its dendrite receives (chemical) synaptic input.

• Note, this is a sub-relation of Relation 4.

Potential names

1. dendrite_innervates
   • con: 'innervates' is used much more generally – and more commonly for tract or nerve to region.
   • pro: easy to name inverse – innervated by dendrite of.
2. dendrite_synapsed_in
   • con: difficult to name inverse
   • pro: it doesn't use innervates, which has other meanings.

Proposed formalization

'N dendrite_synapsed_in some R'expands_to:

  N has_part some ('dendrite ; GO:0030425' that has_part some (
     'pre-synaptic membrane ; GO:0042734' that part_of some (
        'synapse ; GO:0045202' that part_of some R )))

domain: neuron

Rule: If n dendrite_synapsed_in r1 and r1 part_of r2 then n dendrite_synapsed_in r2

Rule Expressed as property chain: dendrite_synapsed_in o part_of -> dendrite_synapsed_in

Examples:

1. Every 'projection neuron DL1 adPN' dendrite_innervates some 'antennal lobe glomerulus DL1'

Relation 8

Relation between neuron and the neuron projection bundle (e.g. a tract, or a bundle in a nerve) that one of its neuron projections (axon, dendrite) travels through.

Potential names (currently preferred name in italics)

1. fasciculates_with
   • inverse relation: has_fasciculating_component ?
2. fasciculates_in
   • inverse relation: has_fasciculating_component ?
3. has_fasciculating_neuron_projection

Proposed formalization

instance level definition: x fasciculates_with y iff* exists some 'neuron projection' (np) and np part_of x and np overlaps y and np aligned_with y

domain: neuron

range: 'neuron projection bundle ; CARO:0001001'

inverse_of: has_fasciculating_neuron

SubPropertyOf: overlaps

Examples:

1. Every 'projection neuron DL1 adPN' fasciculates_with some 'inner antenno-cerebral tract'

Notes:

• As we need to say two things about np, we can't use the class level expansion system to capture this.

Relation 9

Relation between a neuron projection bundle (e.g.- a tract in the brain) and the projection of some neuron that is fasciculated with that bundle.

• Note – it’s likely that this relation will not be used directly as it required named neuron projections (!) but it is critical to defining relation 10.

Proposed names:

has_fasciculating_neuron_projection

Proposed formalization

Instance level defintion: T has_fasciculating_neuron_projection np iff: np overlaps T AND np aligned_with T

Domain: neuron projection bundle ; CARO:0001001' Range: 'neuron projection ; GO0043005' SubPropertyOf: overlaps

Relation 10

Relation between a neuron projection bundle (e.g. a tract) and a region in which one of more of its component neuron projections synapses to targets or receives synaptic input

Potential names:

1. Innervates
   • pro: fits with the most common usage of the term innervates
   • con: If we use it here, we can't use it for Relation 5 (above).
   • inverse: innervated_by

Proposed formalization

'T innervates some R' Expands_to:

  T has_fasciculating_neuron_projection that has_synapse_in some R 

Rule: if X innervates Y and Y is part_of Z then Z innervates Z.

Rule expressed as property chain: innervates o part_of -> part_of

domain: 'neuron projection bundle ; CARO:0001001'

Full expansion (expanding component defs):

  'has fasciculating component' some (
     'neuron projection and (
        'has part' some (
           'postsynaptic membrane' and 'part of' some (
              'synapse' and 'part of' some Y?
           )
        ) 
     ) or (
        'has part' some (
           'presynaptic membrane' and 'part of' some (
              'synapse' and 'part of' some ?Y
           )
        )
     )
  )

Examples:

1. Every inner 'antenno-cerebral tract' innervates some 'lateral horn'
2.  Every inner 'antenno-cerebral tract' innervates some 'antennal lobe'

go back to Relations page