neuralib.atlas.ccf.dataframe.RoiClassifierDataFrame

class neuralib.atlas.ccf.dataframe.RoiClassifierDataFrame[source]

Bases: DataFrameWrapper

RoiClassifierDataFrame with each roi (rows)

Required fields:
- acronym - area acronym
- AP_location: anterior-posterior coordinates (mm)
- DV_location: dorsal-ventral coordinates (mm)
- ML_location: medial-lateral coordinates (mm)
- channel: fluorescence channel (i.e., gfp, rfp, mcherry, …)
- source: source name (i.e., if circuit tracing, you can give source tracing area…)

__init__(df, *, cached_dir=None, invalid_post_processing_cache=False)[source]

Parameters:

df (DataFrame) – DataFrame with required fields
cached_dir (str | Path | PathLike[str] | None) – create cached directory
invalid_post_processing_cache (bool) – invalid post processing cache if cached_dir is not None

Methods

`__init__`(df, *[, cached_dir, ...])
`clear`([n])	See polars.DataFrame.clear.
`clone`()	Clone the wrapper.
`dataframe`([dataframe, may_inplace])	Getter/setter for the internal Polars DataFrame.
`drop`(*columns[, strict])	See polars.DataFrame.drop.
`drop_nulls`(subset)	See polars.DataFrame.drop_nulls.
`fill_nan`([value])	See polars.DataFrame.fill_nan.
`fill_null`([value, strategy, limit])	See polars.DataFrame.fill_null.
`filter`(predicates, *constraints)	See polars.DataFrame.filter.
`filter_capital_name`()	filter only `acronym` contain capital letters
`filter_injection_site`(area, hemisphere)	filter out the injection site labelled
`get_channel_source_dict`()	get channel (key): source (value) dict
`get_classified_column`(field[, strict])	classified column name, with the given
`group_by`(*by[, maintain_order])	See polars.DataFrame.group_by.
`head`([n])	See polars.DataFrame.head.
`join`(other, on[, how, left_on, right_on, ...])	See polars.DataFrame.join.
`lazy`()	Wrap dataframe in a lazy wrapper.
`limit`([n])	See polars.DataFrame.limit.
`partition_by`(by, *more_by[, as_dict])	See polars.DataFrame.partition_by.
`pipe`(function, args, *kwargs)	See polars.DataFrame.pipe.
`post_processing`(*[, filter_capital, tree, ...])	Load the post-processing dataframe
`rename`(mapping)	See polars.DataFrame.rename.
`select`(exprs, *named_exprs)	See polars.DataFrame.select.
`slice`(offset[, length])	See polars.DataFrame.slice.
`sort`(by, *more_by[, descending, nulls_last, ...])	See polars.DataFrame.sort.
`tail`([n])	See polars.DataFrame.tail.
`to_normalized`(norm, level, *[, source, ...])	To the normalized dataframe (example as volume normalized) .
`to_subregion`(region, *[, unit, ...])	To the subregion dataframe (example as Visual region: VIS) .
`with_columns`(exprs, *named_exprs)	See polars.DataFrame.with_columns.
`with_family_columns`()	with column `family` ('HB', 'HY', 'TH', 'MB', 'CB', 'CTXpl', 'HPF', 'ISOCORTEX', 'OLF', 'CTXsp')
`with_hemisphere_column`([invert])	with column `hemisphere` with which hemisphere
`with_overlap_copy`()	Copy overlap channels counts to individual channels, only used if overlapped roi is not counted for individual channels
`with_row_index`([name, offset])	See polars.DataFrame.with_row_index.
`with_tree_columns`()	with customized hierarchical tree structure based on allen brain

Attributes

`channel_counts`	channel counts dataframe
`channels`	list of channel names
`columns`	See polars.DataFrame.columns.
`is_overlapped_channel`	whether there is overlapped channel
`n_channels`	number of channel
`n_sources`	number of source
`schema`	See polars.DataFrame.schema.
`source_counts`	source counts dataframe
`sources`	list of source names

__init__(df, *, cached_dir=None, invalid_post_processing_cache=False)[source]

Parameters:

df (DataFrame) – DataFrame with required fields
cached_dir (str | Path | PathLike[str] | None) – create cached directory
invalid_post_processing_cache (bool) – invalid post processing cache if cached_dir is not None

dataframe(dataframe=None, may_inplace=True)[source]

Getter/setter for the internal Polars DataFrame.

Parameters:

dataframe (DataFrame | None) – Optional new dataframe to set.
may_inplace (bool) – If True, update current instance. Otherwise, return new instance.

Returns:

The current dataframe or a modified wrapper instance.

Return type:

DataFrame | RoiClassifierDataFrame

property channels: list[str]: list of channel names

property n_channels: int: number of channel

property channel_counts: DataFrame: channel counts dataframe

property source_counts: DataFrame: source counts dataframe

property sources: list[str]: list of source names

property n_sources: int: number of source

property is_overlapped_channel: bool: whether there is overlapped channel

get_channel_source_dict()[source]

get channel (key): source (value) dict

Return type:: dict[str, str]

get_classified_column(field, strict=True)[source]

classified column name, with the given

Parameters:

field (int | str | None) – int: tree_[level]; str: get column; None: acronym
strict (bool) – strict check for the pre-defined classified column

Return type:

str

post_processing(*, filter_capital=True, tree=True, family=True, hemisphere=True, copy_overlap=True, filter_injection=None)[source]

Load the post-processing dataframe

Parameters:

filter_capital (bool) – filter only acronym contain capital letters
tree (bool) – with customized hierarchical tree structure based on allen brain
family (bool) – with column family (‘HB’, ‘HY’, ‘TH’, ‘MB’, ‘CB’, ‘CTXpl’, ‘HPF’, ‘ISOCORTEX’, ‘OLF’, ‘CTXsp’
hemisphere (bool) – with column hemisphere with which hemisphere
copy_overlap (bool) – Copy overlap channels counts to individual channels, only set true if overlapped roi is not counted for individual channels
filter_injection (tuple[str, str] | None) – filter out the injection site labelled

Returns:

Return type:

Self

filter_injection_site(area, hemisphere)[source]

filter out the injection site labelled

Parameters:

area (str) – brain area
hemisphere (str) – which hemisphere

Returns:

Return type:

Self

filter_capital_name()[source]

filter only acronym contain capital letters

Return type:: Self

with_tree_columns()[source]

with customized hierarchical tree structure based on allen brain

Return type:: Self

with_family_columns()[source]

with column family (‘HB’, ‘HY’, ‘TH’, ‘MB’, ‘CB’, ‘CTXpl’, ‘HPF’, ‘ISOCORTEX’, ‘OLF’, ‘CTXsp’)

Return type:: Self

with_hemisphere_column(invert=False)[source]

with column hemisphere with which hemisphere

Parameters:: invert (bool) – invert hemisphere. Default ML >= 0 (ipsi), ML < 0 (contra)
Returns:
Return type:: Self

with_overlap_copy()[source]

Copy overlap channels counts to individual channels, only used if overlapped roi is not counted for individual channels

Return type:: Self

to_normalized(norm, level, *, source=None, top_area=None, rest_as_others=False, hemisphere='both', animal=None, volume_norm_backend='cellatlas')[source]

To the normalized dataframe (example as volume normalized)

┌─────────┬────────┬────────┬───────────┬────────────┬────────────────┬────────────┐
│ source  ┆ tree_2 ┆ counts ┆ fraction  ┆ hemisphere ┆ Volumes [mm^3] ┆ normalized │
│ ---     ┆ ---    ┆ ---    ┆ ---       ┆ ---        ┆ ---            ┆ ---        │
│ str     ┆ str    ┆ u32    ┆ f64       ┆ str        ┆ f64            ┆ f64        │
╞═════════╪════════╪════════╪═══════════╪════════════╪════════════════╪════════════╡
│ overlap ┆ ACA    ┆ 1208   ┆ 29.997517 ┆ both       ┆ 5.222484       ┆ 231.307537 │
│ pRSC    ┆ ACA    ┆ 3296   ┆ 22.822324 ┆ both       ┆ 5.222484       ┆ 631.117254 │
│ …       ┆ …      ┆ …      ┆ …         ┆ …          ┆ …              ┆ …          │
│ pRSC    ┆ VIS    ┆ 4035   ┆ 27.939344 ┆ both       ┆ 12.957203      ┆ 311.409797 │
│ overlap ┆ VIS    ┆ 628    ┆ 15.594736 ┆ both       ┆ 12.957203      ┆ 48.46725   │
│ aRSC    ┆ VIS    ┆ 3865   ┆ 12.627005 ┆ both       ┆ 12.957203      ┆ 298.289682 │
└─────────┴────────┴────────┴───────────┴────────────┴────────────────┴────────────┘

Parameters:

norm (Literal['channel', 'volume', 'cell', 'none']) – ROIS_NORM_TYPE
level (int | str | None) – tree level for determine which level of classified column
source (str | None) – filter only the given source
top_area (int | None) – filter only the given top areas (sorted based on fraction)
rest_as_others (bool) – vertical concat a region called other with the roi filtered by top area
hemisphere (Literal['ipsi', 'contra', 'both']) – filter only the given hemisphere
animal (str | None) – with animal id column animal
volume_norm_backend (Literal['cellatlas', 'brainglobe']) – volume normalization backend. {‘cellatlas’, ‘brainglobe’}

Returns:

RoiNormalizedDataFrame

Return type:

RoiNormalizedDataFrame

to_subregion(region, *, unit='fraction', source_order=None, show_col=None, animal=None, normalize=True)[source]

To the subregion dataframe (example as Visual region: VIS)

┌─────────┬───────────┬───────────┬───────────┬───┬──────────┬──────────┬──────────┬──────────┐
│ source  ┆ VISam     ┆ VISp      ┆ VISpm     ┆ … ┆ VISal    ┆ VISpor   ┆ VISli    ┆ VISpl    │
│ ---     ┆ ---       ┆ ---       ┆ ---       ┆   ┆ ---      ┆ ---      ┆ ---      ┆ ---      │
│ str     ┆ f64       ┆ f64       ┆ f64       ┆   ┆ f64      ┆ f64      ┆ f64      ┆ f64      │
╞═════════╪═══════════╪═══════════╪═══════════╪═══╪══════════╪══════════╪══════════╪══════════╡
│ overlap ┆ 39.649682 ┆ 15.127389 ┆ 28.025478 ┆ … ┆ 3.025478 ┆ 2.707006 ┆ 1.592357 ┆ 0.159236 │
│ aRSC    ┆ 32.160414 ┆ 28.952135 ┆ 23.05304  ┆ … ┆ 6.080207 ┆ 1.293661 ┆ 2.069858 ┆ 0.07762  │
│ pRSC    ┆ 25.947955 ┆ 27.95539  ┆ 27.459727 ┆ … ┆ 3.122677 ┆ 2.973978 ┆ 1.982652 ┆ 1.016109 │
└─────────┴───────────┴───────────┴───────────┴───┴──────────┴──────────┴──────────┴──────────┘

Parameters:

region (str) – region name
unit (Literal['counts', 'fraction']) – value unit. {‘counts’, ‘fraction’}. default is ‘fraction’
source_order (tuple[str, ...] | None) – source order in dataframe (rows)
show_col (str | None) – force set show col to which level. use case: if a low level area name is classified and show in high level (i.e., TH).
animal (str | None) – with animal column as subregion dataframe
normalize (bool) – do subregion normalization. True: normalize per source within region (each source sums to 100% within this region), False: normalize per source to total input counts (fraction of total input for that source). Default is True

Returns:

Return type:

RoiSubregionDataFrame