add metabolites concentration from YMDB

add_conc/code/add_conc.m

@@ -0,0 +1,35 @@
+modelWithConc = importModel('../../model/yeast-GEM.xml');
+data = readtable('../data/allConcData.xlsx');
+data = table2cell(data);
+metConc.allConc = cell(length(modelWithConc.metNames), 1);
+metConc.maxConc = cell(length(modelWithConc.metNames), 1);
+metConc.minConc = cell(length(modelWithConc.metNames), 1);
+
+met.name = modelWithConc.metNames;
+met.kegg = {};
+
+for i = 1:length(met.name)
+    idx = find(strcmp('kegg.compound',modelWithConc.metMiriams{i, 1}.name));
+    if idx
+        met.kegg{i, 1} = modelWithConc.metMiriams{i, 1}.value{idx, 1};
+    end
+    clear idx
+end
+
+metWithConc = {};
+for j = 2:870
+    keggId = data{2, j};
+    idx = find(strcmp(keggId, met.kegg));
+    if idx
+        %fprintf(keggId)
+        metWithConc{end+1} = keggId;
+        for id= 1:length(idx)
+            metConc.allConc{idx(id), 1} = data{3, j};
+            metConc.maxConc{idx(id), 1} = data{4, j};
+            metConc.minConc{idx(id), 1} = data{5, j};
+        end
+    end
+end
+fprintf('In total, %d mets have concentration range.', length(metWithConc))
+modelWithConc.metConc = metConc;
+save('../modelWithConc.mat', 'modelWithConc');

add_conc/code/get_YMDB_data.py

@@ -0,0 +1,58 @@
+'''
+@qqlaoxia 20240306
+'''
+
+import argparse
+import requests
+import pandas as pd
+import json
+import os
+import time
+def extract_data(i):
+    cmp='YMDB'+'{:05d}'.format(i)
+# Send an HTTP request to get the content of the web page
+    url = r'https://www.ymdb.ca/compounds/'+cmp
+    response = requests.get(url)
+    web_content = response.text
+    data = json.loads(web_content)
+
+    if isinstance(data, dict):
+        data = [data]
+    extracted_data = []
+    for item in data:
+        row_data = {key: value for key, value in item.items()}
+        extracted_data.append(row_data)
+
+    df = pd.DataFrame(extracted_data)
+    # if not os.path.exists(cmp):
+    #     os.mkdir(cmp)
+    #
+    # # Save the DataFrame to an Excel file
+    # excel_path = './%s/%s.xlsx' %(cmp,cmp)
+    # df.to_excel(excel_path, index=False)
+    # print(f'Data saved to {excel_path}')
+
+
+    # % rearrange data
+    df_transposed = df.T.reset_index()
+    df_transposed.columns = ['Key', 'Content']
+
+    # Save the transposed DataFrame to a new Excel file
+    transposed_excel_path = f'../data/YMDB/{cmp}_transposed.xlsx'
+    df_transposed.to_excel(transposed_excel_path, index=False, header=True)
+    print(f'Transposed data saved to {transposed_excel_path}')
+    time.sleep(5)
+
+
+if __name__ == '__main__':
+    parser = argparse.ArgumentParser()
+    parser.add_argument('-i', help='input number',type=int)
+    args = parser.parse_args()
+    extract_data(args.i)
+    # It is suggested to used parallel arithmetic in matlab.
+    # parfor i=1:16338
+    # system(['python main.py -i ',num2str(i)]);
+    # end
+
+
+

add_conc/code/organise_data.py

@@ -0,0 +1,88 @@
+####################################################################
+# get concentration
+# The concentration range is calculated according to the following logic
+# For example: In YMDB, three conc is recorded for glutamine ('17140.0 ± 1971.0 umol/L', '2500.0 ± 500.0 umol/L', '15000.0 ± 0.0 umol/L')
+# maximal conc: The maximum concentration plus the corresponding error. In this case, is "17140.0 + 1971.0 umol/L"
+# minimal conc: The minimum concentration minus the corresponding error. In this case, is "2500.0 - 500.0 umol/L"
+
+
+import os
+import numpy as np
+import pandas as pd
+import re
+
+
+def get_conc(rawconc):
+    concSplit = rawconc.split('}, ')
+    # unit: umol/L
+    all_conc = []
+    # unit: mol/L
+    maxmin = []
+    for c in concSplit:
+        regex = r'\'([^\']*)\''
+        result = re.findall(regex, c)
+        try:
+            if (result[5] == 'µM') or (result[5] == 'uM') or (result[5] == 'umol/L'):
+                all_conc.append(str(result[3] + ' ± ' + result[7] + ' umol/L'))
+                maxmin.append((float(result[3]) - float(result[7]))*10**(-6))
+                maxmin.append((float(result[3]) + float(result[7]))*10**(-6))
+        except IndexError:
+            pass
+    return all_conc, maxmin
+
+# load data and put them together
+folder_path = r"../data/YMDB"
+fileName = os.listdir(folder_path)
+data = pd.DataFrame()
+for fn in fileName:
+    file_path = os.path.join(folder_path, fn)
+    file = pd.read_excel(file_path)
+    id = file.iloc[0, 1]
+    if id == 'YMDB00001':
+        data.index = list(file.loc[:, 'Key'])[1:]
+    else:
+        pass
+    data.loc[:, id] = list(file.loc[:, 'Content'])[1:]
+    print(fn)
+# data.to_csv(r'../data/allConcData.csv')
+
+# compute maximal conc and minimal conc
+conc = pd.DataFrame(index=['conc', 'maxConc', 'minConc'], columns=data.columns)
+for i in range(len(data.columns)):
+    print(i)
+    rawconc = data.iloc[44, i]
+    all_conc, maxmin = get_conc(rawconc)
+    if len(all_conc) != 0:
+        conc.loc['conc', data.columns[i]] = all_conc
+        conc.loc['maxConc', data.columns[i]] = np.max(maxmin)
+        conc.loc['minConc', data.columns[i]] = np.min(maxmin)
+    else:
+        pass
+
+data = pd.concat([data, conc], axis=0)
+newdata = data.loc[['name', 'chebi_id', 'kegg_id', 'conc', 'maxConc', 'minConc']]
+cleaned_data = newdata.dropna(axis=1, subset=['conc'])
+cleaned_data.to_excel('../data/allConcData.xlsx')
+# # test unit
+#
+# def check_unit(rawconc):
+#     concSplit = rawconc.split('}, ')
+#     unit = []
+#     try:
+#         for c in concSplit:
+#             regex = r'\'([^\']*)\''
+#             result = re.findall(regex, c)
+#             unit.append(result[5])
+#     except:
+#         pass
+#     return unit
+#
+# all_unit= []
+# for i in range(len(data.columns)):
+#     rawconc = data.iloc[44, i]
+#     unit = check_unit(rawconc)
+#     for u in unit:
+#         if u not in all_unit:
+#             all_unit.append(u)
+#             print(data.columns[i])
+#             print(u)