merge_adata
Combine multiple AnnData objects for single-cell RNA sequencing analysis by concatenating along observations or variables with configurable alignment options.
Instructions
merge multiple adata
Input Schema
TableJSON Schema
| Name | Required | Description | Default |
|---|---|---|---|
| axis | No | Which axis to concatenate along. 'obs' or 0 for observations, 'var' or 1 for variables. | obs |
| join | No | How to align values when concatenating. If 'outer', the union of the other axis is taken. If 'inner', the intersection. | inner |
| merge | No | How elements not aligned to the axis being concatenated along are selected. | |
| uns_merge | No | How the elements of .uns are selected. Uses the same set of strategies as the merge argument, except applied recursively. | |
| label | No | label different adata, Column in axis annotation (i.e. .obs or .var) to place batch information in. | |
| keys | No | Names for each object being added. These values are used for column values for label or appended to the index if index_unique is not None. | |
| index_unique | No | Whether to make the index unique by using the keys. If provided, this is the delimiter between '{orig_idx}{index_unique}{key}'. | |
| fill_value | No | When join='outer', this is the value that will be used to fill the introduced indices. | |
| pairwise | No | Whether pairwise elements along the concatenated dimension should be included. |
Implementation Reference
- src/scmcp/tool/util.py:84-88 (handler)The core handler function that merges multiple AnnData objects using anndata.concat, taking adata_dic and kwargs.
def merge_adata(adata_dic, **kwargs): import anndata as ad adata = ad.concat(adata_dic, **kwargs) return adata - src/scmcp/schema/util.py:53-92 (schema)Pydantic model defining the input schema for the merge_adata tool, including parameters like axis, join, merge, etc., passed via kwargs to ad.concat.
class ConcatAdataModel(JSONParsingModel): """Model for concatenating AnnData objects""" axis: Literal['obs', 0, 'var', 1] = Field( default='obs', description="Which axis to concatenate along. 'obs' or 0 for observations, 'var' or 1 for variables." ) join: Literal['inner', 'outer'] = Field( default='inner', description="How to align values when concatenating. If 'outer', the union of the other axis is taken. If 'inner', the intersection." ) merge: Optional[Literal['same', 'unique', 'first', 'only']] = Field( default=None, description="How elements not aligned to the axis being concatenated along are selected." ) uns_merge: Optional[Literal['same', 'unique', 'first', 'only']] = Field( default=None, description="How the elements of .uns are selected. Uses the same set of strategies as the merge argument, except applied recursively." ) label: Optional[str] = Field( default=None, description="label different adata, Column in axis annotation (i.e. .obs or .var) to place batch information in. " ) keys: Optional[List[str]] = Field( default=None, description="Names for each object being added. These values are used for column values for label or appended to the index if index_unique is not None." ) index_unique: Optional[str] = Field( default=None, description="Whether to make the index unique by using the keys. If provided, this is the delimiter between '{orig_idx}{index_unique}{key}'." ) fill_value: Optional[Any] = Field( default=None, description="When join='outer', this is the value that will be used to fill the introduced indices." ) pairwise: bool = Field( default=False, description="Whether pairwise elements along the concatenated dimension should be included." ) - src/scmcp/tool/util.py:40-44 (registration)Registers the merge_adata tool as an MCP types.Tool, specifying name, description, and input schema.
merge_adata_tool = types.Tool( name="merge_adata", description="merge multiple adata", inputSchema=ConcatAdataModel.model_json_schema(), ) - src/scmcp/tool/util.py:98-104 (registration)Registers the merge_adata_tool in the util_tools dictionary for tool dispatching.
util_tools = { "mark_var": mark_var_tool, "list_var": list_var_tool, "list_obs": list_obs_tool, "check_gene": check_gene_tool, "merge_adata": merge_adata_tool, } - src/scmcp/tool/util.py:114-119 (helper)Special handling in run_util_func for merge_adata, which merges all adata_dic and updates the active adata.
if func == "merge_adata": res = merge_adata(ads.adata_dic) ads.adata_dic = {} ads.active = "merge_adata" ads.adata_dic[ads.active] = res else: