Function process_customized

pairwise/prepare_data.py:86–170

()

    #create a dummy synergy dataframe
    def process_customized():
        # drugcomb = process_drugcomb()
        # # drugcomb_colon = drugcomb[drugcomb['tissue_name'] == 'haematopoietic_and_lymphoid']
        # drugcomb_colon = drugcomb.drop_duplicates(subset=['drug1', 'drug2'])

        # # crc_exp = pd.read_csv(os.path.join(ROOT_DIR, 'data', 'cell_line_data','Customized','crc_%s.csv' % "exp"),sep=',')
        # # crc_exp = pd.read_csv(os.path.join(ROOT_DIR, 'data', 'cell_line_data','Customized','tcga_DLBC_%s.csv' % "exp"),sep=',')
        # crc_exp = pd.read_csv(os.path.join(ROOT_DIR, 'data', 'cell_line_data','Customized','orginal_columbia_DLBC_%s_20221014.csv' % "exp"),sep=',')
        # # crc_exp = pd.read_csv(os.path.join(ROOT_DIR, 'data', 'cell_line_data','Customized','lstaudt_DLBC_%s.csv' % "exp"),sep=',')
        # summary_data = pd.DataFrame(columns=['drug1','drug2','cell','tissue_name','score'])
        # cell_list = list(crc_exp.columns)
        # for crc_cell in cell_list: #0:255, 255:
        #     drugcomb_colon['cell'] = crc_cell
        #     drugcomb_colon['score'] = 0
        #     summary_data = summary_data.append(drugcomb_colon, ignore_index=True)

        # summary_data = summary_data.drop_duplicates(subset=['drug1','drug2','cell','tissue_name','score'])
        

        # for dcdb
        dcdb_tmd8 = pd.read_csv(os.path.join(ROOT_DIR, 'data', 'synergy_data','Customized','dcdb_tmd8.csv'),sep=',').iloc[:,1:]
        dcdb_tmd8 = dcdb_tmd8.groupby(['drug1','drug2','cell']).agg({\
            "score":'mean'}).reset_index()
        summary_data = pd.DataFrame(columns=['drug1','drug2','cell','score'])
        #增加az的药物
        azdream = pd.read_csv(os.path.join(ROOT_DIR, 'data', 'synergy_data','Customized','azdream_drug.csv'),sep=',').iloc[:,1:]
        # dcdb_tmd8 = azdream

        # columbia_exp = pd.read_csv(os.path.join(ROOT_DIR, 'data', 'cell_line_data','Customized','orginal_columbia_DLBC_%s_20221014.csv' % "exp"),sep=',')
        columbia_exp = pd.read_csv(os.path.join(ROOT_DIR, 'data', 'cell_line_data','Customized','orginal_lstaudt_DLBC_%s.csv' % "exp"),sep=',')
        cell_list = list(columbia_exp.columns)
        for crc_cell in cell_list:
            dcdb_tmd8['cell'] = crc_cell
            summary_data = summary_data.append(dcdb_tmd8, ignore_index=True)

        summary_data = summary_data.drop_duplicates(subset=['drug1','drug2','cell','score'])
        ## -------------- 所有ccle
        # directory = os.path.join(ROOT_DIR,'data', 'cell_line_data','Customized','TCGA_PAN_RNA_TPM')
        # summary_data = pd.DataFrame(columns=['drug1','drug2','cell','score',"tissue"])
        # counter = 0
        # for file in os.listdir(directory):
        #     filename = os.fsdecode(file)
        #     counter += 1
        #     if filename.endswith("tcga_DLBC_RNA_counts_csv_tpm.csv") and counter <=10:
                
        #         tpm_exp = pd.read_csv(os.path.join(directory, filename))
        #         cell_list = list(tpm_exp.columns)
        #         for crc_cell in cell_list:
        #             dcdb_tmd8['cell'] = crc_cell
        #             dcdb_tmd8['tissue'] = filename.split("_")[1]
        #             summary_data = summary_data.append(dcdb_tmd8, ignore_index=True)
        #         # print(os.path.join(directory, filename))
        # summary_data = summary_data.drop_duplicates(subset=['drug1','drug2','cell','score'])

        # #// for clincials
        # summary_data = pd.DataFrame(columns=['drug1','drug2','cell','score'])
        # clinicals = pd.read_csv(os.path.join(ROOT_DIR, 'data', 'synergy_data','Customized','clinicaltrialsgov_confi.csv'),sep=',').iloc[:,1:]
        # clinicals = clinicals[["compound0_x","compound0_y","cell","score"]].rename(columns={\