marsilea.upset.UpsetData#

class marsilea.upset.UpsetData(data, sets_names=None, items=None, sets_attrs=None, items_attrs=None)#

Bases: object

Handle multiple sets

Normally, the construction methods are used to create a UpsetData

Terminology that you might not be familiar with:

set: a collection of unique items
subset: intersection of sets
degree: The number of sets that intersect with each other
cardinality: The number of items in the subset

Parameters:

databool matrix: A one-hot encode matrix indicates if an item is in a set. Columns are sets and rows are items
sets_namesoptional, array: The name of sets
itemsoptional, array: The name of items
sets_attrsoptional, pd.DataFrame: The attributes of sets, the input index should be the same as names
items_attrsoptional, pd.DataFrame: The attributes of items, the input index should be the same as items

Examples

from marsilea.upset import UpsetData
sets = [[1,2,3,4], [3,4,5,6]]
data = UpsetData.from_sets(sets)

binary_table()#

cardinality()#: The number of items in intersections

degree()#: Intersection between how many sets

filter(min_degree=None, max_degree=None, min_cardinality=None, max_cardinality=None)#

Filter by degree or cardinality

Parameters:

min_degreeint: The minimum degree
max_degreeint: The maximum degree
min_cardinalityint: The minimum cardinality
max_cardinalityint: The maximum cardinality

classmethod from_memberships(items, items_names=None, sets_attrs=None, items_attrs=None)#

Describe the sets an item are in

Parameters:

itemsarray of array of sets_names, dict: The data of items
items_namesoptional: The name of items, if name is not provided, it will be automatically named as “Item 1, Item 2, …”
sets_attrsoptional, pd.DataFrame: The attributes of sets, the input index should be the same as sets_names
items_attrsoptional, pd.DataFrame: The attributes of items, the input index should be the same as items

classmethod from_sets(sets, sets_names=None, sets_attrs=None, items_attrs=None)#

Create UpsetData from a series of sets

Parameters:

setsarray of sets, dict: The sets data
sets_namesoptional: The name of sets, if name is not provided, it will be automatically named as “Set 1, Set 2, …”
sets_attrsoptional, pd.DataFrame: The attributes of sets, the input index should be the same as sets_names
items_attrsoptional, pd.DataFrame: The attributes of items, the input index should be the same as items

get_items_attr(attr)#

Return the attribute of items in the order of plotting

Parameters:

attrkey in items_attrs: Retrieve the attribute of items

Returns:

array of attribute

has_item(item)#: Return a list of sets’ name the item is in

intersection(sets_name)#: Return the items that are shared in different sets

intersection_count()#: The item has occurred in how many sets

property items#

property items_attrs#

mark(present=None, absent=None, min_cardinality=None, max_cardinality=None, min_degree=None, max_degree=None)#

reset()#

property sets_attrs#

property sets_names#

sets_size()#

sets_table()#

sort_sets(ascending=False, order=None)#

Control the order of sets

Parameters:

ascendingbool: Sort in ascending order if True
orderlist: Explicitly specify the order of sets

sort_subsets(by='degree', ascending=False)#

Sort the subsets by degree or cardinality

Parameters:

bystr: Sort by either degree or cardinality
ascendingbool: Sort in ascending order if True