Tutorial 4: VisiumHD dataset
============================

.. raw:: html

    <div style="font-size: 15px;">In this tutorial, we show how to apply scSTADE to identify spatial domains on VisiumHD dataset. As a example, we analyse the VisiumHD dataset.</div>

^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^

.. raw:: html

    <div style="font-size: 15px;">The source code package is freely available at https://github.com/cuiyaxuan/scSTADE/tree/master. The datasets used in this study can be found at https://drive.google.com/drive/folders/1H-ymfCqlDR1wpMRX-bCewAjG5nOrIF51?usp=sharing.</div>

^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^

.. code:: ipython3

    import sys
    sys.path.append("/.../scSTADE-master/scSTADE_Cluster_Functions")
    # Input the path.
    from scSTADE_HD import scSTADE
    import os
    import torch
    import pandas as pd
    import scanpy as sc
    from sklearn import metrics
    import multiprocessing as mp
    
    def setup_seed(seed=41):
        import torch
        import os
        import numpy as np
        import random
        torch.manual_seed(seed)  
        np.random.seed(seed)  # Numpy module.
        random.seed(seed)  # Python random module.
        if torch.cuda.is_available():
            # torch.backends.cudnn.benchmark = False
            torch.backends.cudnn.deterministic = True
            torch.cuda.manual_seed(seed)  
            torch.cuda.manual_seed_all(seed) 
            #os.environ['PYTHONHASHSEED'] = str(seed)
    
    setup_seed(41)
    
    device = torch.device('cpu')
    # call 24 cpu cores
    torch.set_num_threads(24)
    
    
    n_clusters = 20  ###### the number of spatial domains.
    file_fold = '/home/cuiyaxuan/spatialLIBD/square_016um/' #### to your path
    adata = sc.read_visium(file_fold, count_file='filtered_feature_bc_matrix.h5', load_images=True) #### project name
    adata.var_names_make_unique()
    model = scSTADE(adata,device=device,n_top_genes=5000)
    adata = model.train()
    radius = 50
    tool = 'mclust' # mclust, leiden, and louvain
    from utils import clustering
    
    if tool == 'mclust':
       clustering(adata, n_clusters, radius=radius, method=tool, refinement=False)
    elif tool in ['leiden', 'louvain']:
       clustering(adata, n_clusters, radius=radius, method=tool, start=0.1, end=2.0, increment=0.01, refinement=False)
    
    adata.obs['domain']
    adata.obs['domain'].to_csv("label.csv")



.. parsed-literal::

    /home/cuiyaxuan/anaconda3/envs/pytorch/lib/python3.8/site-packages/tqdm/auto.py:22: TqdmWarning: IProgress not found. Please update jupyter and ipywidgets. See https://ipywidgets.readthedocs.io/en/stable/user_install.html
      from .autonotebook import tqdm as notebook_tqdm
    /home/cuiyaxuan/anaconda3/envs/pytorch/lib/python3.8/site-packages/anndata/_core/anndata.py:1830: UserWarning: Variable names are not unique. To make them unique, call `.var_names_make_unique`.
      utils.warn_names_duplicates("var")
    /home/cuiyaxuan/anaconda3/envs/pytorch/lib/python3.8/site-packages/scanpy/readwrite.py:413: DtypeWarning: Columns (1,2,3,4,5) have mixed types. Specify dtype option on import or set low_memory=False.
      positions = pd.read_csv(files['tissue_positions_file'], header=None)


.. image:: 4_Example_VisiumHD_test/human_colon_cancer_domain20-1.png
   :width: 315px
   :height: 277px