HPOSet

An HPOSet is a collection of HPOTerm that can be used to document the clinical information of a patient. At the same time, the phenotypes associated with genes and diseases are also HPOSets. HPOSet can be instantiated in multiple ways, depending on the available data types. Whichever way you choose, you must instantiate the Ontology beforehand.

Instantiation

from_queries(queries)

Instantiate an HPOSet from various inputs

This is the most common way to instantiate HPOSet because it can use all kind of different inputs. Callers must ensure that each query paramater matches a single HPOTerm.

Parameters

queries (list[str or int]) –

str HPO term (e.g.: Scoliosis)
str HPO-ID (e.g.: HP:0002650)
int HPO term id (e.g.: 2650)

Returns

A new HPOSet

Return type

pyhpo.HPOSet

Raises

NameError – Ontology not yet constructed
ValueError – query cannot be converted to HpoTermId
RuntimeError – No HPO term is found for the provided query

Examples

from pyhpo import Ontology
Ontology()
my_set = HPOSet.from_queries([
    "HP:0002650",
    118,
    "Thoracolumbar scoliosis"
])
len(my_set)
# >> 3

from_serialized(pickle)

Instantiate an HPOSet from a serialized HPOSet

This method is used when you have a serialized form of the HPOSet to share between applications. See pyhpo.HPOSet.serialize()

Parameters

pickle (str) – A pickled string of all HPOTerms, e.g. 118+2650

Returns

A new HPOSet

Return type

pyhpo.HPOSet

Raises

NameError – Ontology not yet constructed
ValueError – pickled item cannot be converted to HpoTermId
KeyError – No HPO term is found for the provided query

Examples

from pyhpo import Ontology
Ontology()
my_set = HPOSet.from_serialized("7+118+152+234+271+315+478+479+492+496")
len(my_set
# >> 10

from_gene(gene)

Instantiate an HPOSet from a Gene

Parameters: gene (pyhpo.Gene) – A gene from the ontology
Returns: A new HPOSet
Return type: pyhpo.HPOSet
Raises: NameError – Ontology not yet constructed

Examples

from pyhpo import Ontology
Ontology()
gene_set = HPOSet.from_gene(Ontology.genes[0])
len(gene_set)
# >> 118

from_disease(disease)

Instantiate an HPOSet from an Omim disease

Parameters: gene (pyhpo.Omim) – An Omim disease from the ontology
Returns: A new HPOSet
Return type: pyhpo.HPOSet
Raises: NameError – Ontology not yet constructed

Examples

from pyhpo import Ontology
Ontology()
disease_set = HPOSet.from_disease(Ontology.omim_diseases[0])
len(disease_set)
# >> 18

Instance methods

class HPOSet(terms)

add(term)

Add an HPOTerm to the HPOSet

Parameters

term (HPOTerm or int) – The term to add, either as actual HPOTerm or the integer representation

Raises

NameError – Ontology not yet constructed
KeyError – (only when int are used as input): HPOTerm does not exist

Examples

from pyhpo import Ontology, HPOSet
Ontology()
my_set = HPOSet([])
my_set.add(Ontology[118])
len(my_set) # >> 1
my_set.add(2650)
len(my_set) # >> 2

all_genes()

Returns a set of associated genes

Returns: The union of genes associated with terms in the HPOSet
Return type: set[pyhpo.Gene]
Raises: NameError – Ontology not yet constructed

Examples

from pyhpo import Ontology
Ontology()
disease = list(Ontology.omim_diseases)[0]
for gene in disease.all_genes():
    print(gene.name)

child_nodes()

Returns a new HPOSet that does not contain ancestor terms

If a set contains HPOTerms that are ancestors of other terms in the set, they will be removed. This method is useful to create a set that contains only the most specific terms.

Returns: A new HPOSet that contains only the most specific terms
Return type: pyhpo.HPOSet
Raises: NameError – Ontology not yet constructed

Examples

from pyhpo import Ontology, HPOSet

my_set = HPOSet.from_queries([
    'HP:0002650',
    'HP:0010674',
    'HP:0000925',
    'HP:0009121'
])

child_set = my_set.child_nodes()

len(my_set) # >> 4
len(child_set) # >> 1

information_content(kind='omim')

Returns basic information content stats about the HPOTerms within the set

Parameters

kind (str, default: omim) – Which kind of information content should be calculated. Options are [‘omim’, ‘gene’]

Returns

Dict with the following items

mean - float - Mean information content
max - float - Maximum information content value
total - float - Sum of all information content values
all - list of float - List with all information content values

Return type

dict

Raises

NameError – Ontology not yet constructed

Examples

omim_diseases()

Returns a set of associated diseases

Returns: The union of Omim diseases associated with terms in the HPOSet
Return type: set[pyhpo.Omim]
Raises: NameError – Ontology not yet constructed

Examples

from pyhpo import Ontology
Ontology()
gene = list(Ontology.genes)[0]
for disease in disease.omim_diseases():
    print(disease.name)

remove_modifier()

Returns a new HPOSet that does not contain any modifier terms

This method removes all terms that are not children of HP:0000118 | Phenotypic abnormality

Returns: A new HPOSet that contains only phenotype terms
Return type: pyhpo.HPOSet
Raises: NameError – Ontology not yet constructed

Examples

from pyhpo import Ontology, HPOSet

my_set = HPOSet.from_queries([
    'HP:0002650',
    'HP:0010674',
    'HP:0000925',
    'HP:0009121',
    'HP:0012823',
])

pheno_set = my_set.remove_modifier()

len(my_set) # >> 5
len(pheno_set) # >> 4

replace_obsolete()

Returns a new HPOSet that replaces all obsolete terms with their replacement

If an obsolete term has a replacement term defined it will be replaced, otherwise it will be removed.

Returns: A new HPOSet that contains only phenotype terms
Return type: pyhpo.HPOSet
Raises: NameError – Ontology not yet constructed

Examples

from pyhpo import Ontology, HPOSet

my_set = HPOSet.from_queries([
    'HP:0002650',
    'HP:0010674',
    'HP:0000925',
    'HP:0009121',
    'HP:0410003',
])

active_set = my_set.replace_obsolete()

len(my_set) # >> 5
len(active_set) # >> 5

Ontology[410003] in my_set
# >> True

Ontology[410003] in active_set
# >> False

serialize()

Returns a serialized string representing the HPOSet

Returns: A serialized string uniquely representing the HPOSet, e.g.: 3+118+2650`
Return type: str

Examples

from pyhpo import Ontology
Ontology()
gene_sets = [g.hpo_set() for g in Ontology.genes]
gene_sets[0].serialize()
# >> 7+118+152+234+271+315+478+479+492+496.....

similarity(other, kind, method, combine)

Calculate similarity between this and another HPOSet

This method runs parallelized on all avaible CPU

Parameters

other (pyhpo.HPOSet) – The HPOSet to calculate the similarity to
kind (str, default: omim) –
Which kind of information content to use for similarity calculation

Available options:
- omim
- gene
method (str, default graphic) –
The method to use to calculate the similarity.

Available options:
- resnik - Resnik P, Proceedings of the 14th IJCAI, (1995)
- lin - Lin D, Proceedings of the 15th ICML, (1998)
- jc - Jiang J, Conrath D, ROCLING X, (1997) This is different to PyHPO
- jc2 - Jiang J, Conrath D, ROCLING X, (1997) Same as jc, but kept for backwards compatibility
- rel - Relevance measure - Schlicker A, et.al., BMC Bioinformatics, (2006)
- ic - Information coefficient - Li B, et. al., arXiv, (2010)
- graphic - Graph based Information coefficient - Deng Y, et. al., PLoS One, (2015)
- dist - Distance between terms
combine (str, default funSimAvg) –
The method to combine individual term similarity to HPOSet similarities.

Available options:
- funSimAvg
- funSimMax
- BMA

Returns

Similarity scores

Return type

float

Raises

NameError – Ontology not yet constructed
AttributeError – Invalid kind
RuntimeError – Invalid method or combine

Examples

from pyhpo import Ontology
Ontology()
gene_sets = [g.hpo_set() for g in Ontology.genes]
gene_sets[0].similarity(gene_sets[1])
# >> 0.29546087980270386

similarity_scores(other, kind, method, combine)

Calculate similarity between this HPOSet and a list of other HPOSet

This method runs parallelized on all avaible CPU

Parameters

other (list[pyhpo.HPOSet]) – Calculate similarity between self and every provided HPOSet
kind (str, default: omim) –
Which kind of information content to use for similarity calculation

Available options:
- omim
- gene
method (str, default graphic) –
The method to use to calculate the similarity.

Available options:
- resnik - Resnik P, Proceedings of the 14th IJCAI, (1995)
- lin - Lin D, Proceedings of the 15th ICML, (1998)
- jc - Jiang J, Conrath D, ROCLING X, (1997) This is different to PyHPO
- jc2 - Jiang J, Conrath D, ROCLING X, (1997) Same as jc, but kept for backwards compatibility
- rel - Relevance measure - Schlicker A, et.al., BMC Bioinformatics, (2006)
- ic - Information coefficient - Li B, et. al., arXiv, (2010)
- graphic - Graph based Information coefficient - Deng Y, et. al., PLoS One, (2015)
- dist - Distance between terms
combine (str, default funSimAvg) –
The method to combine individual term similarity to HPOSet similarities.

Available options:
- funSimAvg
- funSimMax
- BMA

Returns

Similarity scores for every comparison

Return type

list[float]

Raises

NameError – Ontology not yet constructed
KeyError – Invalid kind
RuntimeError – Invalid method or combine

Examples

from pyhpo import Ontology
Ontology()
gene_sets = [g.hpo_set() for g in Ontology.genes]
similarities = gene_sets[0].similarity_scores(gene_sets)
similarities[0:4]
# >> [1.0, 0.5000048279762268, 0.29546087980270386, 0.5000059008598328]

terms()

Returns the HPOTerms in the set

Returns: A list of every term in the set
Return type: list[pyhpo.HPOTerm]

Important

The return type of this method will very likely change into an Iterator of HPOTerm. (Info about likely API changes)

Raises

NameError – Ontology not yet constructed
KeyError – No HPO term is found for the provided query

Examples

from pyhpo import Ontology
Ontology()
my_set = list(Ontology.genes)[0].hpo_set()
for term in my_set.terms():
    print(term.name)

toJSON(verbose)

Returns a dict/JSON representation the HPOSet

Parameters

verbose (bool) – Indicates if each HPOTerm should contain verbose information see pyhpo.HpoTerm.toJSON()

Returns

Dict representation of all HPOTerms in the set that can be used for JSON serialization

Return type

Dict

Raises

NameError – Ontology not yet constructed
KeyError – No HPO term is found for the provided query

Examples

from pyhpo import Ontology
Ontology()
my_set = HPOSet.from_serialized("7+118+152+234+271+315+478+479+492+496")
my_set.toJSON()
# >> [
# >>     {'name': 'Autosomal recessive inheritance', 'id': 'HP:0000007', 'int': 7},
# >>     {'name': 'Phenotypic abnormality', 'id': 'HP:0000118', 'int': 118},
# >>     {'name': 'Abnormality of head or neck', 'id': 'HP:0000152', 'int': 152},
# >>     {'name': 'Abnormality of the head', 'id': 'HP:0000234', 'int': 234},
# >>     {'name': 'Abnormality of the face', 'id': 'HP:0000271', 'int': 271},
# >>     {'name': 'Abnormality of the orbital region', 'id': 'HP:0000315', 'int': 315},
# >>     {'name': 'Abnormality of the eye', 'id': 'HP:0000478', 'int': 478},
# >>     {'name': 'Abnormal retinal morphology', 'id': 'HP:0000479', 'int': 479},
# >>     {'name': 'Abnormal eyelid morphology', 'id': 'HP:0000492', 'int': 492},
# >>     {'name': 'Abnormality of eye movement', 'id': 'HP:0000496', 'int': 496}
# >> ]

Not yet implemented

The following instance methods are not yet implemented for pyhpo.HPOSet

variance(self, /)

Calculates the distances between all its term-pairs. It also provides basic calculations for variances among the pairs.

Returns

Tuple with the variance metrices

float Average distance between pairs
int Smallest distance between pairs
int Largest distance between pairs
list of int List of all distances between pairs

Return type

tuple of (int, int, int, list of int)

combinations(self, /)

Helper generator function that returns all possible two-pair combination between all its terms

This function is direction dependent. That means that every pair will appear twice. Once for each direction

BasicHPOSet

A BasicHPOSet is like a normal pyhpo.HPOSet, but:

only child terms are retained, non-specific parent terms are removed
a obsolete terms are replaced or removed
all modifier terms are removed

HPOPhenoSet

A BasicHPOSet is like a normal pyhpo.HPOSet, but:

a obsolete terms are replaced or removed
all modifier terms are removed

Term	HPOSet	BasicHPOSet	HPOPhenoSet
obsolete	✅	❌	❌
modifier	✅	❌	❌
parents	✅	❌	✅