stereo.core.ms_data.MSData

class stereo.core.ms_data.MSData(_data_list=<factory>, _merged_data=None, _names=<factory>, _obs=None, _var=None, _var_type='intersect', _relationship='other', _relationship_info=<factory>, _name_dict=<factory>, _data_dict=<factory>, _MSDataStruct__idx_generator=-1, _scopes_data=<factory>)

MSData is a composite structure of several StereoExpData organized by some relationship.

Parameters:

• data_list (List[StereoExpData] stereo_exp_data array) – An array of stereo_exp_data organized by some relationship defined by _relationship and _relationship_info

• merged_data (stereo_exp_data object) – An stereo_exp_data merged with data_list used batches integrate.

• names (List[str] stereo_exp_data array’s names) – An array of `stereo_exp_data`s’ unique names.

• obs (pandas.DataFrame = None) – pandas.DataFrame describes all the cells or bins observed, indexes mean cell names or bin names, columns mean some math statistic or types produced by bio-information algorithm.

• var (pd.DataFrame = None) – pandas.DataFrame describes genes, similar to _obs.

• _var_type (str = 'intersect') – Which claims that _var is intersected by lots of genes from different samples.

• relationship (str = 'other') – Relationship about samples in _data_list.

• _relationship_info (object) – Relationship extra info.

• tl (object) –

  MSData algorithms collections, include all tools methods inherited from stereo_exp_data and multi-samples

  methods. Methods from stereo_exp_data will organized with mutilthreads while running.

• plt (object) – MSData plot methods collections, same as tl.

Examples

Constructing MSData from two `stereo_exp_data`s.

>>> from stereo.io.reader import read_gef
>>> from stereo.core.ms_data import MSData
>>> data1 = read_gef("../demo_data/SS200000135TL_D1/SS200000135TL_D1.gef")
>>> data2 = read_gef("../demo_data/SS200000135TL_D1/SS200000135TL_D1.tissue.gef")
>>> ms_data = MSData(_data_list=[data1, data2], _names=['raw', 'tissue'], _relationship='other', _var_type='intersect') # noqa
>>> ms_data

ms_data: {‘raw’: (9004, 25523), ‘tissue’: (9111, 20816)} num_slice: 2 names: [‘raw’, ‘tissue’] obs: [‘test_obs_1’] var: [‘test_var_1’] relationship: other var_type: intersect to 20760 tl.result: defaultdict(<class ‘list’>, {})

Constructing MSData one by one using add method.

>>> from stereo.core.ms_data import MSData
>>> from stereo.io.reader import read_gef
>>> ms_data = MSData()
>>> ms_data += read_gef("../demo_data/SS200000135TL_D1/SS200000135TL_D1.gef")
>>> ms_data += read_gef("../demo_data/SS200000135TL_D1/SS200000135TL_D1.tissue.gef")
>>> ms_data
ms_data: {'0': (9004, 25523), '1': (9111, 20816)}
num_slice: 2
names: ['0', '1']
obs: ['test_obs_1']
var: ['test_var_1']
relationship: other
var_type: intersect to 20760
tl.result: defaultdict(<class 'list'>, {})

Slice features like python list.

>>> from stereo.core.ms_data import MSData
>>> from stereo.io.reader import read_gef
>>> ms_data = MSData()
>>> ms_data += read_gef("../demo_data/SS200000135TL_D1/SS200000135TL_D1.gef")
>>> ms_data += read_gef("../demo_data/SS200000135TL_D1/SS200000135TL_D1.tissue.gef")
>>> ms_data += read_gef("../demo_data/SS200000135TL_D1/SS200000135TL_D1.gef")
>>> ms_data += read_gef("../demo_data/SS200000135TL_D1/SS200000135TL_D1.tissue.gef")
>>> ms_data += read_gef("../demo_data/SS200000135TL_D1/SS200000135TL_D1.gef")
>>> ms_data += read_gef("../demo_data/SS200000135TL_D1/SS200000135TL_D1.tissue.gef")
>>> ms_data[3:]
_MSDataView(_names=['3', '4', '5'], _data_list=[StereoExpData object with n_cells X n_genes = 9111 X 20816
bin_type: bins
bin_size: 100
offset_x = 0
offset_y = 2
cells: ['cell_name']
genes: ['gene_name'], ...)

Slice features like DataFrame.

>>> ms_data[('1', '4'):]
_MSDataView(_names=['1', '4'], _data_list=[StereoExpData object with n_cells X n_genes = 9111 X 20816
bin_type: bins
bin_size: 100
offset_x = 0
offset_y = 2
cells: ['cell_name']
genes: ['gene_name'], ...)

__init__(_data_list=<factory>, _merged_data=None, _names=<factory>, _obs=None, _var=None, _var_type='intersect', _relationship='other', _relationship_info=<factory>, _name_dict=<factory>, _data_dict=<factory>, _MSDataStruct__idx_generator=-1, _scopes_data=<factory>)

Methods

__init__([_data_list, _merged_data, _names, ...])
add_data([data, names])	rtype: object
del_data(name)
generate_scope_key([scope])
get_data_list(key_idx_list)
integrate([scope, remove_existed])	Integrate some single-samples specified by scope to a merged one.
remove_scopes_data(scope)
rename(mapper)	rtype: object
reset_name([start_idx, default_key])	rtype: object
reset_position([mode])
split_after_batching_integrate()
to_integrate(scope, res_key, _from[, type, ...])	Integrate an obs column or a var column from some single-samples spcified by _from to the merged sample.
to_isolated(scope, res_key, to[, type, ...])	Copy a result from mss group specfied by scope to some single-samples specfied by to.
to_msdata(data[, batch_key, relationship, ...])
var_percent()

Attributes

data_list
merged_data
mss
names
num_slice
obs
plt
relationship
relationship_info
scopes_data
shape
tl
var
var_type