neuralib.atlas.ccf.classifier

neuralib.atlas.ccf.classifier.FluorReprType: alias of dict[str, str]

typeddict neuralib.atlas.ccf.classifier.UserInjectionConfig[source]

typing.TypedDict.

Required Keys:

area (str) – injection area
hemisphere (Literal['ipsi', 'contra', 'both']) – injection hemisphere
ignore (bool) – whether local roi counts will be ignored
fluor_repr (dict[str, str]) – fluorescence color and tracing source alias pairs

class neuralib.atlas.ccf.classifier.RoiClassifier[source]

__init__(ccf_dir, merge_level=None, plane='coronal', config=None)[source]

Classifier for see the ROIs distribution across the mouse brain

Parameters:

ccf_dir (AbstractCCFDir) – ccf directory info
merge_level (int | str | None) – which number (int) of hierarchical merge area, or the col name (str) for output generate
plane (PLANE_TYPE) – slice cutting orientation
config (UserInjectionConfig | None) – UserInjectionConfig for customized injection info

ccf_dir

merge_level

plane

ignore_injection_site: Final[bool]

fluor_repr: FluorReprType

property has_overlap: bool: If has overlap channel counts

property parsed_df: DataFrame

dataframe after parsing

Example:

┌───────────────────────────────────┬─────────┬─────────────┬─────────────┬─────────────┬─────────┬─────────┬────────┬───────────────────────────┬──────────────┬────────┬────────────┬────────────┬────────────┬────────────┬────────────┬───────────┐
│ name                              ┆ acronym ┆ AP_location ┆ DV_location ┆ ML_location ┆ avIndex ┆ channel ┆ source ┆ abbr                      ┆ acronym_abbr ┆ hemi.  ┆ merge_ac_0 ┆ merge_ac_1 ┆ merge_ac_2 ┆ merge_ac_3 ┆ merge_ac_4 ┆ family    │
│ ---                               ┆ ---     ┆ ---         ┆ ---         ┆ ---         ┆ ---     ┆ ---     ┆ ---    ┆ ---                       ┆ ---          ┆ ---    ┆ ---        ┆ ---        ┆ ---        ┆ ---        ┆ ---        ┆ ---       │
│ str                               ┆ str     ┆ f64         ┆ f64         ┆ f64         ┆ i64     ┆ str     ┆ str    ┆ str                       ┆ str          ┆ str    ┆ str        ┆ str        ┆ str        ┆ str        ┆ str        ┆ str       │
╞═══════════════════════════════════╪═════════╪═════════════╪═════════════╪═════════════╪═════════╪═════════╪════════╪═══════════════════════════╪══════════════╪════════╪════════════╪════════════╪════════════╪════════════╪════════════╪═══════════╡
│ Ectorhinal area/Layer 5           ┆ ECT5    ┆ -3.03       ┆ 4.34        ┆ -4.5        ┆ 377     ┆ gfp     ┆ aRSC   ┆ Ectorhinal area           ┆ ECT          ┆ contra ┆ ECT        ┆ ECT        ┆ ECT        ┆ ECT        ┆ ECT        ┆ ISOCORTEX │
│ Perirhinal area layer 6a          ┆ PERI6a  ┆ -3.03       ┆ 4.42        ┆ -4.37       ┆ 372     ┆ gfp     ┆ aRSC   ┆ Perirhinal area           ┆ PERI         ┆ contra ┆ PERI       ┆ PERI       ┆ PERI       ┆ PERI       ┆ PERI       ┆ ISOCORTEX │
│ …                                 ┆ …       ┆ …           ┆ …           ┆ …           ┆ …       ┆ …       ┆ …      ┆ …                         ┆ …            ┆ …      ┆ …          ┆ …          ┆ …          ┆ …          ┆ …          ┆ …         │
│ Ventral auditory area layer 6a    ┆ AUDv6a  ┆ -2.91       ┆ 3.52        ┆ 4.46        ┆ 156     ┆ rfp     ┆ pRSC   ┆ Ventral auditory area     ┆ AUDv         ┆ ipsi   ┆ AUD        ┆ AUD        ┆ AUD        ┆ AUD        ┆ AUDv       ┆ ISOCORTEX │
│ Ectorhinal area/Layer 6a          ┆ ECT6a   ┆ -2.91       ┆ 4.14        ┆ 4.47        ┆ 378     ┆ rfp     ┆ pRSC   ┆ Ectorhinal area           ┆ ECT          ┆ ipsi   ┆ ECT        ┆ ECT        ┆ ECT        ┆ ECT        ┆ ECT        ┆ ISOCORTEX │
│ Temporal association areas layer… ┆ TEa5    ┆ -2.91       ┆ 4.02        ┆ 4.55        ┆ 365     ┆ rfp     ┆ pRSC   ┆ Temporal association area ┆ TEa          ┆ ipsi   ┆ TEa        ┆ TEa        ┆ TEa        ┆ TEa        ┆ TEa        ┆ ISOCORTEX │
└───────────────────────────────────┴─────────┴─────────────┴─────────────┴─────────────┴─────────┴─────────┴────────┴───────────────────────────┴──────────────┴────────┴────────────┴────────────┴────────────┴────────────┴────────────┴───────────┘

property classified_column: Literal['acronym_abbr', 'merge_ac_0', 'merge_ac_1', 'merge_ac_2', 'merge_ac_3', 'merge_ac_4', 'family']

property channels: list[str]: list of fluorescence channels (sorted based on fluor_repr)

property n_channels: int: number of fluorescence channels

property sources: list[str]: list of unique sources

property n_sources: int: number of source name

property areas: list[str]: list of area in the given classified_column

property n_total_rois: int: number of total rois

property nroi_channel: DataFrame: channels and counts Dataframe

property nroi_source: DataFrame: channels and counts Dataframe

property nroi_area: DataFrame: dict with key (area name) and value (number of cells)

get_percent_sorted_df(hemisphere='both')[source]

Get percentage-sorted dataframe

Parameters:: hemisphere (Literal['ipsi', 'contra', 'both']) – which hemisphere {‘ipsi’, ‘contra’, ‘both’}
Returns:: dataframe
Return type:: DataFrame

hemi: optional col. if hemisphere is`both`, then sum together

Example of merge level equal to 2:

┌─────────┬────────────┬────────┬────────┬───────────┐
│ source  ┆ merge_ac_2 ┆ *hemi. ┆ n_rois ┆ percent   │
│ ---     ┆ ---        ┆ ---    ┆ ---    ┆ ---       │
│ str     ┆ str        ┆ str    ┆ u32    ┆ f64       │
╞═════════╪════════════╪════════╪════════╪═══════════╡
│ pRSC    ┆ VIS        ┆ ipsi   ┆ 5701   ┆ 22.761209 │
│ aRSC    ┆ MO         ┆ ipsi   ┆ 2277   ┆ 21.360225 │
│ overlap ┆ MO         ┆ ipsi   ┆ 571    ┆ 17.799252 │
│ aRSC    ┆ ACA        ┆ ipsi   ┆ 1729   ┆ 16.219512 │
│ overlap ┆ SUB        ┆ ipsi   ┆ 494    ┆ 15.399002 │
│ …       ┆ …          ┆ …      ┆ …      ┆ …         │
│ pRSC    ┆ MED        ┆ contra ┆ 1      ┆ 0.003992  │
│ pRSC    ┆ PPN        ┆ contra ┆ 1      ┆ 0.003992  │
│ pRSC    ┆ APr        ┆ ipsi   ┆ 1      ┆ 0.003992  │
│ pRSC    ┆ BMA        ┆ contra ┆ 1      ┆ 0.003992  │
│ pRSC    ┆ P5         ┆ contra ┆ 1      ┆ 0.003992  │
└─────────┴────────────┴────────┴────────┴───────────┘

get_classified_data(norm=None, *, top_area=None, source=None, add_other=False, supply_overlap=True, hemisphere='both', area=None)[source]

processed data for plotting / visualization

Parameters:

norm (MouseBrainRoiNormHandler | None) – BrainMapNormHandler
top_area (int | None) – select top ranks area based on channel-based normalized percentage
source (Source | None) – specify source, if not then produce all source
add_other (bool) – the rest of regions (after top selection), classified as other (i.e., pie chart)
supply_overlap (bool) – add overlap roi counts into other channel(s)
hemisphere (HEMISPHERE_TYPE) – filter the output data with selected hemisphere {‘ipsi’, ‘contra’, ‘both’}
area (Area | list[Area] | None) – filter the output data with selected area

Returns:

RoiClassifiedNormTable

Return type:

RoiClassifiedNormTable

class neuralib.atlas.ccf.classifier.RoiClassifiedNormTable[source]

source_classifier: RoiClassifier

norm_type: ROIS_NORM_TYPE | None

hemisphere: HEMISPHERE_TYPE

data: pl.DataFrame

processed data after ROIClassifier

Optional col: hemi. , only existed if init with either hemisphere is ‘ipsi’ or ‘contra’
<ROIS_NORM_TYPE>_norm depending on norm_type

Example of data with volume normalization method:

┌─────────┬────────────┬────────┬───────────┬───┬───────────┬───────────┬─────────────────┬────────┐
│ channel ┆ merge_ac_2 ┆ n_rois ┆ percent   ┆ … ┆ Volumes   ┆ n_neurons ┆ *volume_norm_n_r┆ animal │
│ ---     ┆ ---        ┆ ---    ┆ ---       ┆   ┆ [mm^3]    ┆ ---       ┆ ois             ┆ ---    │
│ str     ┆ str        ┆ i64    ┆ f64       ┆   ┆ ---       ┆ i64       ┆ ---             ┆ str    │
│         ┆            ┆        ┆           ┆   ┆ f64       ┆           ┆ f64             ┆        │
╞═════════╪════════════╪════════╪═══════════╪═══╪═══════════╪═══════════╪═════════════════╪════════╡
│ overlap ┆ ACA        ┆ 423    ┆ 30.344333 ┆ … ┆ 5.222484  ┆ 337372    ┆ 80.995934       ┆ YW051  │
│ gfp     ┆ MO         ┆ 3352   ┆ 24.545987 ┆ … ┆ 22.248234 ┆ 985411    ┆ 150.663641      ┆ YW051  │
│ rfp     ┆ ACA        ┆ 1383   ┆ 23.791502 ┆ … ┆ 5.222484  ┆ 337372    ┆ 264.816494      ┆ YW051  │
│ gfp     ┆ ACA        ┆ 3130   ┆ 22.920328 ┆ … ┆ 5.222484  ┆ 337372    ┆ 599.331616      ┆ YW051  │
│ …       ┆ …          ┆ …      ┆ …         ┆ … ┆ …         ┆ …         ┆ …               ┆ …      │
│ overlap ┆ SS         ┆ 1      ┆ 0.071736  ┆ … ┆ 37.177937 ┆ 2384622   ┆ 0.026898        ┆ YW051  │
│ overlap ┆ ECT        ┆ 1      ┆ 0.071736  ┆ … ┆ 3.457703  ┆ 387378    ┆ 0.289209        ┆ YW051  │
│ overlap ┆ TEa        ┆ 1      ┆ 0.071736  ┆ … ┆ 3.860953  ┆ 386396    ┆ 0.259003        ┆ YW051  │
│ rfp     ┆ TT         ┆ 1      ┆ 0.017203  ┆ … ┆ 1.734078  ┆ 124596    ┆ 0.576675        ┆ YW051  │
└─────────┴────────────┴────────┴───────────┴───┴───────────┴───────────┴─────────────────┴────────┘

property value_col: str: column name for the value

property classified_column: Literal['acronym_abbr', 'merge_ac_0', 'merge_ac_1', 'merge_ac_2', 'merge_ac_3', 'merge_ac_4', 'family']: classified column based on the RoiClassifier

with_areas(areas)[source]

filter the data with selected area

Parameters:: areas (Area | list[Area]) – area or list of area
Returns:: RoiClassifiedNormTable
Return type:: Self

with_hemisphere(hemi)[source]

filter the data with selected hemisphere

Parameters:: hemi (Literal['ipsi', 'contra', 'both']) – {‘ipsi’, ‘contra’}
Returns:: RoiClassifiedNormTable
Return type:: Self

get_bias_value_dataframe(yunit='percent', to_index=True, verbose=True)[source]

Dataframe with bias value using either

simple subtraction using either percent or n_rois in self.data (bias_value)
index calculation (bias index)

Parameters:

yunit (str)
to_index – to bias index (1og2 (P_areaA / P_areaB)). refer to Chen et al., 2022. biorxiv. otherwise, do simple subtraction NOTE that index is currently calculate based on channel-normalization value (percent col)
verbose – do print for output

Returns:

DataFrame

Return type:

DataFrame

Example:

┌────────────┬────────────┐
│ merge_ac_1 ┆ bias_value*│
│ ---        ┆ ---        │
│ str        ┆ f64        │
╞════════════╪════════════╡
│ MO         ┆ -18.591053 │
│ SS         ┆ -5.362179  │
│ …          ┆ …          │
│ CP         ┆ 3.897736   │
│ VIS        ┆ 12.899966  │
└────────────┴────────────┘

to_bokeh(yunit='percent')[source]

to bokeh data ColumnSource input structure

Parameters:

yunit (str) – {‘percent’, ‘n_rois’}

Returns:

key: x & counts

value: (area, channel) & neuronal counts or percentage

Return type:

dict[str, list]

to_winner_dataframe()[source]

For ternary plot in plotly module

Returns:: dataframe
Return type:: DataFrame

┌────────────┬─────────┬──────┬──────┬───────┬────────┐
│ merge_ac_1 ┆ overlap ┆ pRSC ┆ aRSC ┆ total ┆ winner │
│ ---        ┆ ---     ┆ ---  ┆ ---  ┆ ---   ┆ ---    │
│ str        ┆ i64     ┆ i64  ┆ i64  ┆ i64   ┆ str    │
╞════════════╪═════════╪══════╪══════╪═══════╪════════╡
│ ACA        ┆ 1000    ┆ 1048 ┆ 2865 ┆ 4913  ┆ gfp    │
│ RSP        ┆ 327     ┆ 1086 ┆ 1345 ┆ 2758  ┆ gfp    │
│ …          ┆ …       ┆ …    ┆ …    ┆ …     ┆ …      │
│ CB         ┆ 0       ┆ 0    ┆ 1    ┆ 1     ┆ gfp    │
│ CUN        ┆ 0       ┆ 0    ┆ 1    ┆ 1     ┆ gfp    │
└────────────┴─────────┴──────┴──────┴───────┴────────┘

write_csv(out, animal_id_col=None)[source]

Parameters:

out (Path)
animal_id_col (str | None)

__init__(source_classifier, norm_type, hemisphere, data)

Method generated by attrs for class RoiClassifiedNormTable.

Parameters:

source_classifier (RoiClassifier)
norm_type (ROIS_NORM_TYPE | None)
hemisphere (HEMISPHERE_TYPE)
data (pl.DataFrame)

Return type:

None

neuralib.atlas.ccf.classifier.supply_overlap_dataframe(df, supply_channel_name)[source]

Supply overlap counting in the parsed dataframe.

** Only used if excluding the overlap cell while counting the rfp and gfp channel

Parameters:

df (DataFrame) – dataframe contain source column with overlap literal in the cell
supply_channel_name (Iterable[str]) – channel names to be supplied from overlap

Return type:

DataFrame

neuralib.atlas.ccf.classifier.parse_csv(ccf_dir, fluor_repr, *, plane='coronal', df=None, invert_hemi=False)[source]

Narrow down the info in the ccf roi output

columns abbr, acronym_abbr and hemi are added to the df

Parameters:

ccf_dir (AbstractCCFDir | None) – if None, directly give df arg
fluor_repr (FluorReprType) – FluorReprType
plane (PLANE_TYPE) – PLANE_TYPE {‘coronal’, ‘sagittal’, ‘transverse’}
df (pl.DataFrame | None) – if None, give ccf_dir arg
invert_hemi (bool) – if True, then ML_location >= 0 is ipsilateral site. otherwise, is contralateral site

Returns:

DataFrame

Return type:

pl.DataFrame