Part II *ESAI calculating*
---------------------------------

With the output of ``SEVtras.sEV_recognizer`` in :doc:`/Part I sEVs recognizing` and cell matrix with cell type, SEVtras can track each sEV to original cell type and calculate sEV secretion activity index (ESAI).

**Note1:** The input cell matrix should contain sample and cell type information in the ``obs`` of adata. 

**Note2:** This command can be compatible to all platform, including Jupyter on Windows. If you encounter a gesapy file occupation error on Windows platform, please try to rerun the same command. 

SEVtras provides function ``ESAI_calculator`` to evaluate dynamic of cellular sEV secretion activity. 

.. code-block:: python

    import SEVtras
    SEVtras.ESAI_calculator(adata_ev_path='./tests/sEV_SEVtras.h5ad', adata_cell_path='./tests/test_cell.h5ad', out_path='./outputs', Xraw=False, OBSsample='batch', OBScelltype='celltype')

The first two parameters represent the path to sEV- and cell- anndata objects. sEV-anndata object comes from the output of SEVtras.sEV_recognizer. The cell-anndata object is the same as conventional single cell analysis data coming from ``filtered_feature_bc_matrix`` directory in Cell Ranger **outs**. 

The third parameter specifies the path of the ``ESAI_calculator`` outputs. The outputs include:  

-  an adata file combining both ``adata_ev`` and ``adata_cell``, named *SEVtras_combined.h5ad*; **ESAI_c** and **ESAI_cS** are indexed in the obs matrix, and represent the sEV secretion activity index at the cell type level across all samples and resolved sample by sample; 
-  two csv files calculating the sEV secretion activity index at the sample level (**ESAI**) and cell type level (**ESAI_cS**), named  *ESAI_sample.csv* and *ESAI_celltype.csv*\; 
-  one pdf file embedding sEVs and cells in a umap , named *SEVumap.pdf*\; 
-  and two pdf files plotting the **ESAI_c** in a umap, named *ESAIumap.pdf* and *ESAIumap_sample.pdf*\; Here, **ESAI_c** is the sEV secretion activity at the cell type level in all your samples, and **ESAI_cS** means the sEV secretion activity at the cell type level resolved sample by sample. 

The fourth parameter means whether to use the raw object in the ``adata_cell`` or not. If ``adata_cell`` has been filtered or normalized, please set ``Xraw=True``, and ``adata_cell.raw`` will be used (Note: save raw ``adata_cell`` as ``adata_cell.raw`` before filtering). 

The last two parameters define which index represents the sample and cell type information in the ``obs`` of adata. By default, SEVtras uses the index of ``batch`` and ``celltype`` in the ``obs`` of ``adata_cell``\. We can change the index with the parameters and *OBSsample* and *OBScelltype*\. 

**Note:** The sample information in ``adata_ev`` is in the key of ``batch`` by default. If *OBSsample* != ``batch``, please change the key in the ``adata_ev`` too before v0.2.12.

The original cell type for each droplet listed in the ``obsm`` of ``SEVtras_sEVs.h5ad`` indexed as ``source``. 

The result of *SEVumap.pdf* and *ESAIumap.pdf* is similar to the following: 

.. image:: ./Part2.png
    :width: 600px
    :align: center
    
A higher level of ESAI  indicates that this cell type has intense sEV secretion activity, which may be related to tumor tumor malignancy, invasion, metastasis and other disease progression. We recommend performing association analysis of ESAI with clinical indicators.