[Python] How to load image data with JupyterLab in UBK-RAP

I have buied a project which contains fundus images in UK-Biobank. And I want to load them in UKB-RAP with JupyterLab so that I can use the images to do my work. How to achieve it?

For example, I know the entity is participant, I know the name is p21015_i0_a0, I know the eid is 1234567, and I know the folder_path is Assessment centre/Eye measures/Retinal optical coherence tomography. How to load it to JupyterLab?

The JupyterLab is provided by "DNAnexus Platform", I want to load image from UKB to it.

Some informations about p21015_i0_a0 are attached, maybe useful.



Additionally, I know a way to load phenotype data, as follows. But how to load bulk(image) data with JupyterLab?

#Step 1. Get the file containing all available data-fields in your Dataset by running the following command:
import dxpy
import subprocess

# Automatically discover dispensed dataset ID and load the dataset
dispensed_dataset_id = dxpy.find_one_data_object(typename='Dataset', name='app*.dataset', folder='/', name_mode='glob')['id']

# Get project ID
project_id = dxpy.find_one_project()["id"]
dataset = (':').join([project_id, dispensed_dataset_id])

cmd = ["dx", "extract_dataset", dataset, "-ddd", "--delimiter", ","]
subprocess.check_call(cmd)
#The above command will generate 3 *.csv files, and the *.dataset.data_dictionary.csv file contains full information about all available data-fields (the field names are in the second column).


#Step 2. Collect all data-field names of interest, and write them into a variable field_names_protein. Here is an example collecting all proteins.
import os
import glob

path = os.getcwd()

data_dict_csv = glob.glob(os.path.join(path, "*.data_dictionary.csv"))[0]
data_dict_df = pd.read_csv(data_dict_csv)
data_dict_df.head()

field_names = list(
data_dict_df.loc[data_dict_df["entity"] == "olink_instance_0", "name"].values
)

field_names_str = [f"olink_instance_0.{f}" for f in field_names]
field_names_protein = ",".join(field_names_str)


#Step 3. Extract the data using dx extract_dataset command.
cmd = ['dx',
'extract_dataset',
dataset,
'--fields',
field_names_protein,
'--delimiter',
',',
'--output',
<filename.csv>,
]
subprocess.check_call(cmd)


Any suggestions will be useful for me. Thanks.

Continue reading...