mkmd.sh doc-gen

apis/python/doc/annotation_dataframe.md

@@ -55,46 +55,59 @@ def keys() -> List[str]
 Returns the column names for the `obs` or `var` dataframe.  For obs and varp, `.keys()` is a
 keystroke-saver for the more general array-schema accessor `attr_names`.
 
+<a id="tiledbsc.annotation_dataframe.AnnotationDataFrame.keyset"></a>
+
+#### keyset
+
+```python
+def keyset() -> Set[str]
+```
+
+Same as `.keys` but returns as set.
+
 <a id="tiledbsc.annotation_dataframe.AnnotationDataFrame.dim_select"></a>
 
 #### dim\_select
 
 ```python
-def dim_select(ids)
+def dim_select(ids, attrs=None)
 ```
 
-Selects a slice out of the dataframe with specified `obs_ids` (for `obs`) or `var_ids` (for `var`).
-If `ids` is `None`, the entire dataframe is returned.
+Selects a slice out of the dataframe with specified `obs_ids` (for `obs`) or `var_ids` (for
+`var`).  If `ids` is `None`, the entire dataframe is returned.  Similarly, if `attrs` are
+provided, they're used for the query; else, all attributes are returned.
 
 <a id="tiledbsc.annotation_dataframe.AnnotationDataFrame.df"></a>
 
 #### df
 
 ```python
-def df(ids=None) -> pd.DataFrame
+def df(ids=None, attrs=None) -> pd.DataFrame
 ```
 
 Keystroke-saving alias for `.dim_select()`. If `ids` are provided, they're used
-to subselect; if not, the entire dataframe is returned.
+to subselect; if not, the entire dataframe is returned. If `attrs` are provided,
+they're used for the query; else, all attributes are returned.
 
 <a id="tiledbsc.annotation_dataframe.AnnotationDataFrame.attribute_filter"></a>
 
 #### attribute\_filter
 
 ```python
-def attribute_filter(query_string, col_names_to_keep)
+def attribute_filter(query_string, attrs=None)
 ```
 
-Selects from obs/var using a TileDB-Py `QueryCondition` string such as
-`cell_type == "blood"`. Returns None if the slice is empty.
-This is a v1 implementation for the prototype/demo timeframe.
+Selects from obs/var using a TileDB-Py `QueryCondition` string such as `cell_type ==
+"blood"`.  If `attrs` is `None`, returns all column names in the dataframe; use `[]` for
+`attrs` to select none of them.  Any column names specified in the `query_string` must be
+included in `attrs` if `attrs` is not `None`.  Returns `None` if the slice is empty.
 
 <a id="tiledbsc.annotation_dataframe.AnnotationDataFrame.from_dataframe"></a>
 
 #### from\_dataframe
 
 ```python
-def from_dataframe(dataframe: pd.DataFrame, extent: int) -> None
+def from_dataframe(dataframe: pd.DataFrame, extent: int = 2048) -> None
 ```
 
 Populates the `obs` or `var` subgroup for a SOMA object.
@@ -104,14 +117,3 @@ Populates the `obs` or `var` subgroup for a SOMA object.
 - `dataframe`: `anndata.obs`, `anndata.var`, `anndata.raw.var`.
 - `extent`: TileDB `extent` parameter for the array schema.
 
-<a id="tiledbsc.annotation_dataframe.AnnotationDataFrame.to_dataframe"></a>
-
-#### to\_dataframe
-
-```python
-def to_dataframe() -> pd.DataFrame
-```
-
-Reads the TileDB `obs` or `var` array and returns a type of pandas dataframe
-and dimension values.
-

apis/python/doc/assay_matrix.md

@@ -62,7 +62,7 @@ Note: currently implemented via data scan -- will be optimized for TileDB core 2
 #### dim\_select
 
 ```python
-def dim_select(obs_ids, var_ids)
+def dim_select(obs_ids, var_ids) -> pd.DataFrame
 ```
 
 Selects a slice out of the matrix with specified `obs_ids` and/or `var_ids`.
@@ -80,6 +80,26 @@ def df(obs_ids=None, var_ids=None) -> pd.DataFrame
 Keystroke-saving alias for `.dim_select()`. If either of `obs_ids` or `var_ids`
 are provided, they're used to subselect; if not, the entire dataframe is returned.
 
+<a id="tiledbsc.assay_matrix.AssayMatrix.csr"></a>
+
+#### csr
+
+```python
+def csr(obs_ids=None, var_ids=None) -> scipy.sparse.csr_matrix
+```
+
+Like `.df()` but returns results in `scipy.sparse.csr_matrix` format.
+
+<a id="tiledbsc.assay_matrix.AssayMatrix.csc"></a>
+
+#### csc
+
+```python
+def csc(obs_ids=None, var_ids=None) -> scipy.sparse.csc_matrix
+```
+
+Like `.df()` but returns results in `scipy.sparse.csc_matrix` format.
+
 <a id="tiledbsc.assay_matrix.AssayMatrix.from_matrix_and_dim_values"></a>
 
 #### from\_matrix\_and\_dim\_values
@@ -91,6 +111,10 @@ def from_matrix_and_dim_values(matrix, row_names, col_names) -> None
 Imports a matrix -- nominally `scipy.sparse.csr_matrix` or `numpy.ndarray` -- into a TileDB
 array which is used for `X`, `raw.X`, `obsp` members, and `varp` members.
 
+The `row_names` and `col_names` are row and column labels for the matrix; the matrix may be
+`scipy.sparse.csr_matrix`, `scipy.sparse.csc_matrix`, `numpy.ndarray`, etc.
+For ingest from `AnnData`, these should be `ann.obs_names` and `ann.var_names`.
+
 <a id="tiledbsc.assay_matrix.AssayMatrix.ingest_data_whole"></a>
 
 #### ingest\_data\_whole

apis/python/doc/assay_matrix_group.md

@@ -86,12 +86,15 @@ def __contains__(name)
 
 Implements the `in` operator, e.g. `"data" in soma.X`.
 
-<a id="tiledbsc.assay_matrix_group.AssayMatrixGroup.from_matrix_and_dim_values"></a>
+<a id="tiledbsc.assay_matrix_group.AssayMatrixGroup.add_layer_from_matrix_and_dim_values"></a>
 
-#### from\_matrix\_and\_dim\_values
+#### add\_layer\_from\_matrix\_and\_dim\_values
 
 ```python
-def from_matrix_and_dim_values(matrix, row_names, col_names) -> None
+def add_layer_from_matrix_and_dim_values(matrix,
+                                         row_names: str,
+                                         col_names: str,
+                                         layer_name="data") -> None
 ```
 
 Populates the `X` or `raw.X` subgroup for a `SOMA` object.  For `X` and `raw.X`, nominally `row_names` will be `anndata.obs_names` and `col_names` will be `anndata.var_names` or `anndata.raw.var_names`.  For `obsp` elements, both will be `anndata.obs_names`; for `varp elements, both will be `anndata.var_names`.

apis/python/doc/soma.md

@@ -87,3 +87,35 @@ def var_keys()
 
 An alias for `soma.var.ids()`.
 
+<a id="tiledbsc.soma.SOMA.cell_count"></a>
+
+#### cell\_count
+
+```python
+def cell_count() -> int
+```
+
+Returns the `obs_id` in `soma.obs`.
+
+<a id="tiledbsc.soma.SOMA.get_obs_value_counts"></a>
+
+#### get\_obs\_value\_counts
+
+```python
+def get_obs_value_counts(obs_label: str) -> pd.DataFrame
+```
+
+Given an obs label, e.g. `cell_type`, returns a dataframe count the number of different
+values for that label in the SOMA.
+
+<a id="tiledbsc.soma.SOMA.get_var_value_counts"></a>
+
+#### get\_var\_value\_counts
+
+```python
+def get_var_value_counts(var_label: str) -> pd.DataFrame
+```
+
+Given an var label, e.g. `feature_name`, returns a dataframe count the number of different
+values for that label in the SOMA.
+

apis/python/doc/soma_collection.md

@@ -33,6 +33,16 @@ Create a new `SOMACollection` object. The existing group is opened at the specif
 - `uri`: URI of the TileDB group
 - `verbose`: Print status messages
 
+<a id="tiledbsc.soma_collection.SOMACollection.__len__"></a>
+
+#### \_\_len\_\_
+
+```python
+def __len__() -> int
+```
+
+Implements `len(soco)`. Returns the number of elements in the collection.
+
 <a id="tiledbsc.soma_collection.SOMACollection.add"></a>
 
 #### add
@@ -53,6 +63,16 @@ def remove(soma: SOMA) -> None
 
 Removes a `SOMA` from the `SOMACollection`.
 
+<a id="tiledbsc.soma_collection.SOMACollection.keys"></a>
+
+#### keys
+
+```python
+def keys() -> None
+```
+
+Returns the names of the SOMAs in the collection.
+
 <a id="tiledbsc.soma_collection.SOMACollection.__iter__"></a>
 
 #### \_\_iter\_\_
@@ -84,3 +104,59 @@ def __getitem__(name)
 Returns a `SOMA` element at the given name within the group, or `None` if no such
 member exists.  Overloads the `[...]` operator.
 
+<a id="tiledbsc.soma_collection.SOMACollection.cell_count"></a>
+
+#### cell\_count
+
+```python
+def cell_count() -> int
+```
+
+Returns sum of `soma.cell_count()` over SOMAs in the collection.
+
+<a id="tiledbsc.soma_collection.SOMACollection.find_unique_obs_values"></a>
+
+#### find\_unique\_obs\_values
+
+```python
+def find_unique_obs_values(obs_label: str)
+```
+
+Given an `obs` label such as `cell_type` or `tissue`, returns a list of unique values for
+that label among all SOMAs in the collection.
+
+<a id="tiledbsc.soma_collection.SOMACollection.find_unique_var_values"></a>
+
+#### find\_unique\_var\_values
+
+```python
+def find_unique_var_values(var_label: str)
+```
+
+Given an `var` label such as `feature_name`, returns a list of unique values for
+that label among all SOMAs in the collection.
+
+<a id="tiledbsc.soma_collection.SOMACollection.get_obs_value_counts"></a>
+
+#### get\_obs\_value\_counts
+
+```python
+def get_obs_value_counts(obs_label: str, do_sum: bool)
+```
+
+For a given obs label, e.g. "cell_type", count the number of occurrences of different values in
+SOMAs in the collection. If `do_sum` is false, count the number of SOMAs having that value. If
+`do_sum` is true, count the total number of instances of that value across the collection.
+
+<a id="tiledbsc.soma_collection.SOMACollection.get_var_value_counts"></a>
+
+#### get\_var\_value\_counts
+
+```python
+def get_var_value_counts(var_label: str, do_sum: bool)
+```
+
+For a given var label, e.g. "feature_name", count the number of occurrences of different values in
+SOMAs in the collection. If `do_sum` is false, count the number of SOMAs having that value. If
+`do_sum` is true, count the total number of instances of that value across the collection.
+

apis/python/doc/tiledb_array.md

@@ -96,6 +96,16 @@ def has_attr_name(attr_name: str) -> bool
 
 Returns true if the array has the specified attribute name, false otherwise.
 
+<a id="tiledbsc.tiledb_array.TileDBArray.has_attr_names"></a>
+
+#### has\_attr\_names
+
+```python
+def has_attr_names(attr_names: List[str]) -> bool
+```
+
+Returns true if the array has all of the specified attribute names, false otherwise.
+
 <a id="tiledbsc.tiledb_array.TileDBArray.show_metadata"></a>
 
 #### show\_metadata

apis/python/doc/util.md

@@ -2,6 +2,17 @@
 
 # tiledbsc.util
 
+<a id="tiledbsc.util.is_local_path"></a>
+
+#### is\_local\_path
+
+```python
+def is_local_path(path: str) -> bool
+```
+
+Returns information about start time of an event. Nominally float seconds since the epoch,
+but articulated here as being compatible with the format_elapsed function.
+
 <a id="tiledbsc.util.get_start_stamp"></a>
 
 #### get\_start\_stamp
@@ -24,6 +35,28 @@ def format_elapsed(start_stamp, message: str)
 Returns the message along with an elapsed-time indicator, with end time relative to start
 start from `get_start_stamp`. Used for annotating elapsed time of a task.
 
+<a id="tiledbsc.util.X_and_ids_to_sparse_matrix"></a>
+
+#### X\_and\_ids\_to\_sparse\_matrix
+
+```python
+def X_and_ids_to_sparse_matrix(
+    Xdf: pd.DataFrame,
+    row_dim_name: str,
+    col_dim_name: str,
+    attr_name: str,
+    row_labels: List[str],
+    col_labels: List[str],
+    return_as: str = "csr"
+) -> Union[scipy.sparse.csr_matrix, scipy.sparse.csc_matrix]
+```
+
+This is needed when we read a TileDB X.df[:]. Since TileDB X is sparse 2D string-dimensioned,
+the return value of which is a dict with three columns -- obs_id, var_id, and value. For
+conversion to anndata, we need make a sparse COO/IJV-format array where the indices are
+not strings but ints, matching the obs and var labels.
+The `return_as` parameter must be one of `"csr"` or `"csc"`.
+
 <a id="tiledbsc.util.ETATracker"></a>
 
 ## ETATracker Objects