main.py

Dockerfile

@@ -5,18 +5,17 @@ RUN echo "downloading basic packages for installation"
 RUN apt-get update
 RUN apt-get install -y tmux wget curl nano less git
 
-WORKDIR /home/
+WORKDIR /src/
 
-COPY requirements_docker.txt ./
+COPY requirements.txt ./
 
-RUN pip install --no-cache-dir -r requirements_docker.txt
-RUN git clone https://github.com/NiklasTR/petri.git
-RUN pip install petri/python
+RUN pip install --no-cache-dir -r requirements.txt
+# RUN git clone https://github.com/NiklasTR/petri.git
+# RUN pip install petri/python
 
 COPY . .
 
-
 # run a test
-RUN python predict.py
+# RUN python predict.py
 # CMD ["uvicorn", "main:app", "--host", "0.0.0.0", "--port", "80"]
-#CMD ["bash"]
+CMD ["bash"]

main.py

@@ -0,0 +1,110 @@
+import sys 
+import inference_VS_2
+import os
+import yaml
+import fastapi # a package we use to receive and return results via an API
+
+# define the API and a base message
+app = FastAPI()
+@app.get("/")
+def root():    
+    return "waiting for input"
+
+# below we define the key function used within a docker and 
+# TODO define API endpoint?
+# @app.get("/{input}")
+def predict(
+    protein: str, # a PDB protein structure file
+    small_molecule_library: str, # "an SDF file containing >=2 small molecule ligands")
+    ):
+    # custom changes
+    args = inference_VS_2.parse_arguments()
+    args = args[0]
+    args.inference_path = '/src/tmp' # copy the input files to this directory for renaming and processing
+    args.output_directory = '/outputs'
+
+    # formatting input
+    protein = str(protein)
+    small_molecule_library = str(small_molecule_library)     
+
+    # isolate the argument path basenames
+    protein_base = os.path.basename(protein)
+    small_molecule_library_base = os.path.basename(small_molecule_library)
+
+    # defining file name
+    protein_destination = args.inference_path + '/dummy/protein_' + protein_base
+    small_molecule_library_destination = args.inference_path + '/dummy/ligands_' + small_molecule_library_base
+
+    # moving files from the paths defined in the arguments to the input directory for processing
+    os.system('mkdir -p ' + args.inference_path + '/dummy')
+    os.system('mv ' + protein + ' ' + protein_destination)
+    os.system('mv ' + small_molecule_library + ' ' + small_molecule_library_destination)
+
+    # the dataurl go package does not like .sdf files, the input should be given in .txt - something to add to petri
+    #small_molecule_library_sdf = os.path.splitext(small_molecule_library_destination)[0]+'.sdf'
+    #print("converting library to sdf: " + small_molecule_library_sdf)
+    #os.system('mv ' + small_molecule_library_destination + ' ' + small_molecule_library_sdf)
+
+    # adding missings args, only works for one run_dir
+    args.multi_ligand = True
+    # args.model_parameters['noise_initial'] = 0
+
+    # running the inference - this is the main function - lifted from __main__ in inference_VS_2.py
+    if args.config:
+        config_dict = yaml.load(args.config, Loader=yaml.FullLoader)
+        arg_dict = args.__dict__
+        for key, value in config_dict.items():
+            if isinstance(value, list):
+                for v in value:
+                    arg_dict[key].append(v)
+            # dropping comparisson with CMD line arguments
+            #else:
+            #    if key in cmdline_args:
+            #        continue
+            #    arg_dict[key] = value
+        args.config = args.config.name
+    else:
+        config_dict = {}
+
+
+    for run_dir in args.run_dirs:
+        args.checkpoint = f'runs/{run_dir}/best_checkpoint.pt'
+        config_dict['checkpoint'] = f'runs/{run_dir}/best_checkpoint.pt'
+        # overwrite args with args from checkpoint except for the args that were contained in the config file
+        arg_dict = args.__dict__
+        with open(os.path.join(os.path.dirname(args.checkpoint), 'train_arguments.yaml'), 'r') as arg_file:
+            checkpoint_dict = yaml.load(arg_file, Loader=yaml.FullLoader)
+        for key, value in checkpoint_dict.items():
+            if (key not in config_dict.keys()):
+                if isinstance(value, list):
+                    for v in value:
+                        arg_dict[key].append(v)
+                else:
+                    arg_dict[key] = value
+        args.model_parameters['noise_initial'] = 0
+        if args.inference_path == None:
+            inference_VS_2.inference(args)
+        elif args.multi_ligand == True:
+            print('Running Multi-Ligand')
+            #print(args)
+            inference_VS_2.multi_lig_inference(args)
+        else:
+            inference_VS_2.inference_from_files(args)
+
+    # moving the output file to the output directory
+    ouput_name = os.listdir(args.output_directory + '/dummy')[0]
+    output_path_sdf = args.output_directory + '/dummy/' + ouput_name
+    print(output_path_sdf)
+
+    # the dataurl go package does not like .sdf files, the input should be given in .txt - something to add to petri
+    # output_path = os.path.splitext(output_path_sdf)[0]+'.txt'
+    # print("converting library to txt: " + output_path)
+    # os.system('mv ' + output_path_sdf + ' ' + output_path)
+
+    # output
+    # output = Path(output_path)
+    return(output_path_sdf)
+
+if __name__ == "__main__":
+    #sys.argv[1]
+    print(predict(sys.argv[1], sys.argv[2]))