Download Data¶

We recommend the user visit our DISCO website and filter the sample based on the metadata provided on the website.

In this tutorial, we will provide a quick guideline for downloading the data using discotoolkit. Steps:

• Visit the website to find a sample or cell type of interest.
• Utilize the dt.filter_disco_metadata function to filter the data from the database.
• Employ the dt.download_disco_data function to download the data in the h5ad extension based on the filtered sample.

In [1]:

# for google colab
# pip install discotoolkit
# first import the installed package
import discotoolkit as dt
import os
import scanpy as sc
import anndata as ad
import re
import matplotlib.pyplot as plt

# adding ignore warning to clean the code
import warnings

# Ignore all warnings
warnings.filterwarnings('ignore')

# for google colab # pip install discotoolkit # first import the installed package import discotoolkit as dt import os import scanpy as sc import anndata as ad import re import matplotlib.pyplot as plt # adding ignore warning to clean the code import warnings # Ignore all warnings warnings.filterwarnings('ignore')

In [2]:

# setting params for the visualisation
# verbosity: errors (0), warnings (1), info (2), hints (3)
sc.settings.verbosity = 1
sc.settings.set_figure_params(
    dpi=300, frameon=False, figsize=(3, 3), facecolor='white')

# Set the plotting backend to 'umap-learn' for interactive plot
sc.settings.plotting_backend = 'umap-learn'

# setting params for the visualisation # verbosity: errors (0), warnings (1), info (2), hints (3) sc.settings.verbosity = 1 sc.settings.set_figure_params( dpi=300, frameon=False, figsize=(3, 3), facecolor='white') # Set the plotting backend to 'umap-learn' for interactive plot sc.settings.plotting_backend = 'umap-learn'

For this example, we wanted to download LYVE1 macrophage from lung tissue as our cell type of interest. We subset only the highly confident cell type prediction based on our manual annotation and the minimum 300 cells per sample. Notice that we specify include_cell_type_children as true to include all sub-cell types if applicable.

In [3]:

# filtering the sample based on the metadata
filter = dt.Filter(
    sample_id = None,
    project_id = None,
    tissue = "lung",
    platform = ["10x3'"],
    sample_type = ["control"],
    cell_type = "LYVE1 macrophage",
    cell_type_confidence = "high",
    include_cell_type_children = True,
    min_cell_per_sample = 300
)

# filtering the sample based on the metadata filter = dt.Filter( sample_id = None, project_id = None, tissue = "lung", platform = ["10x3'"], sample_type = ["control"], cell_type = "LYVE1 macrophage", cell_type_confidence = "high", include_cell_type_children = True, min_cell_per_sample = 300 )

In [4]:

# apply the filter of interest
metadata = dt.filter_disco_metadata(filter)

# apply the filter of interest metadata = dt.filter_disco_metadata(filter)

INFO:root:Filtering sample
INFO:root:Retrieving ontology from DISCO database
INFO:root:13 samples and 7074 cells were found

After applying the filter to the database, now we can download the dataset into the disco_data directory.

In [5]:

# download data to the local directory
dir_path = "disco_data"
download_log = dt.download_disco_data(metadata, output_dir = dir_path)

# download data to the local directory dir_path = "disco_data" download_log = dt.download_disco_data(metadata, output_dir = dir_path)

100%|██████████| 11.5M/11.5M [00:00<00:00, 46.7MB/s]
... storing 'cell_type' as categorical
... storing 'sample_id' as categorical
... storing 'project_id' as categorical
... storing 'feature_types' as categorical
... storing 'genome' as categorical
100%|██████████| 8.23M/8.23M [00:00<00:00, 47.8MB/s]
... storing 'cell_type' as categorical
... storing 'sample_id' as categorical
... storing 'project_id' as categorical
... storing 'feature_types' as categorical
... storing 'genome' as categorical
100%|██████████| 3.11M/3.11M [00:00<00:00, 34.3MB/s]
... storing 'cell_type' as categorical
... storing 'sample_id' as categorical
... storing 'project_id' as categorical
... storing 'feature_types' as categorical
... storing 'genome' as categorical
100%|██████████| 4.86M/4.86M [00:00<00:00, 12.8MB/s]
... storing 'cell_type' as categorical
... storing 'sample_id' as categorical
... storing 'project_id' as categorical
... storing 'feature_types' as categorical
... storing 'genome' as categorical
100%|██████████| 43.6M/43.6M [00:00<00:00, 53.4MB/s]
... storing 'cell_type' as categorical
... storing 'sample_id' as categorical
... storing 'project_id' as categorical
... storing 'feature_types' as categorical
... storing 'genome' as categorical
100%|██████████| 51.4M/51.4M [00:01<00:00, 49.5MB/s]
... storing 'cell_type' as categorical
... storing 'sample_id' as categorical
... storing 'project_id' as categorical
... storing 'feature_types' as categorical
... storing 'genome' as categorical
100%|██████████| 51.9M/51.9M [00:04<00:00, 13.6MB/s]
... storing 'cell_type' as categorical
... storing 'sample_id' as categorical
... storing 'project_id' as categorical
... storing 'feature_types' as categorical
... storing 'genome' as categorical
100%|██████████| 55.7M/55.7M [00:01<00:00, 45.7MB/s]
... storing 'cell_type' as categorical
... storing 'sample_id' as categorical
... storing 'project_id' as categorical
... storing 'feature_types' as categorical
... storing 'genome' as categorical
100%|██████████| 53.6M/53.6M [00:01<00:00, 39.9MB/s]
... storing 'cell_type' as categorical
... storing 'sample_id' as categorical
... storing 'project_id' as categorical
... storing 'feature_types' as categorical
... storing 'genome' as categorical
100%|██████████| 58.4M/58.4M [00:03<00:00, 17.7MB/s]
... storing 'cell_type' as categorical
... storing 'sample_id' as categorical
... storing 'project_id' as categorical
... storing 'feature_types' as categorical
... storing 'genome' as categorical
100%|██████████| 51.9M/51.9M [00:01<00:00, 46.7MB/s]
... storing 'cell_type' as categorical
... storing 'sample_id' as categorical
... storing 'project_id' as categorical
... storing 'feature_types' as categorical
... storing 'genome' as categorical
100%|██████████| 57.4M/57.4M [00:01<00:00, 49.2MB/s]
... storing 'cell_type' as categorical
... storing 'sample_id' as categorical
... storing 'project_id' as categorical
... storing 'feature_types' as categorical
... storing 'genome' as categorical
100%|██████████| 48.9M/48.9M [00:01<00:00, 37.9MB/s]
... storing 'cell_type' as categorical
... storing 'sample_id' as categorical
... storing 'project_id' as categorical
... storing 'feature_types' as categorical
... storing 'genome' as categorical

We can check to see if the files have been download successfully using os library with function listdir()

In [6]:

# double checking to see if the files have been download successf
os.listdir(dir_path)

# double checking to see if the files have been download successf os.listdir(dir_path)

Out[6]:

['GSM5068927.h5ad',
 'GSM3891621.h5ad',
 'GSM3891625.h5ad',
 'GSM4827184.h5ad',
 'GSM5068933.h5ad',
 'GSM5068936.h5ad',
 'GSM4827185.h5ad',
 'GSM7017456.h5ad',
 'GSM5068932.h5ad',
 'GSM5068934.h5ad',
 'GSM5068928.h5ad',
 'GSM5068924.h5ad',
 'GSM5068929.h5ad']

Since the data has been successfully downloaded, we can then use the scanpy library to visualize the UMAP of the downloaded data. This can be achieved by concatenating the files in the downloaded directory into one combined anndata. The combined anndata is useful for downstream analysis or integration, enabling further study of the scRNA-seq.

In [7]:

directory = 'disco_data'
adata_list = []

# Read and store datasets
for filename in os.listdir(directory):
    if filename.endswith('.h5ad'):
        filepath = os.path.join(directory, filename)
        adata = sc.read(filepath)

        # Rename columns with periods in `.obs` attribute
        for col in adata.obs.columns:
            new_col = re.sub(r'\.', '_', col)
            adata.obs.rename(columns={col: new_col}, inplace=True)

        # Rename columns with periods in `.var` attribute
        for col in adata.var.columns:
            new_col = re.sub(r'\.', '_', col)
            adata.var.rename(columns={col: new_col}, inplace=True)

        del adata.raw

        adata_list.append(adata)
        
# Concatenate the datasets
combined_adata = ad.concat(adata_list)

# making the obs name unique
combined_adata.obs_names_make_unique()

# Print the concatenated dataset
print(combined_adata)

directory = 'disco_data' adata_list = [] # Read and store datasets for filename in os.listdir(directory): if filename.endswith('.h5ad'): filepath = os.path.join(directory, filename) adata = sc.read(filepath) # Rename columns with periods in `.obs` attribute for col in adata.obs.columns: new_col = re.sub(r'\.', '_', col) adata.obs.rename(columns={col: new_col}, inplace=True) # Rename columns with periods in `.var` attribute for col in adata.var.columns: new_col = re.sub(r'\.', '_', col) adata.var.rename(columns={col: new_col}, inplace=True) del adata.raw adata_list.append(adata) # Concatenate the datasets combined_adata = ad.concat(adata_list) # making the obs name unique combined_adata.obs_names_make_unique() # Print the concatenated dataset print(combined_adata)

AnnData object with n_obs × n_vars = 7074 × 33538
    obs: 'cell_type', 'sample_id', 'project_id'

Great! We have successfully filtered and downloaded the cell type-specific data from the DISCO database.

---

Last update: October 24, 2024
Created: May 11, 2023