Function ALL

pairwise/select_features.py:80–96

()

        return selected_genes

    def ALL():
        # use before batch corrected CCLE most vairance genes
        # index is drug, so we need clean up the df
        temp = pd.read_csv(os.path.join(ROOT_DIR, 'data', 'cell_line_data','CCLE','CCLE_exp.csv'),sep=',')
        temp.columns = ['Entrez gene id']+[split_it_cell(_) for _ in list(temp.columns)[1:]]
        df_transpose = temp.T
        df_transpose.columns = df_transpose.iloc[0]
        processed_data = df_transpose.drop(df_transpose.index[0])

        var_df = processed_data.var(axis=1)
        selected_genes = list(var_df.sort_values(ascending=False).iloc[:1000].index)
        with open(os.path.join(ROOT_DIR, 'data', 'drug_data','input_drug_data.npy'), 'rb') as file:
            data_dicts = np.load(file, allow_pickle=True).item()

        drug_target = list(data_dicts['drug_target'].index)

        return list(set(selected_genes+drug_target))

    # select genes based on criterion (variance or STRING)
    function_mapping = {'variance':'filter_by_variance', 'STRING':'filter_by_706_genes', 'dti':'filter_by_2000_genes',\