Omics Dashboard Python Client

A python client for accessing an Omics Dashboard service.

Installation

Requirements

Python 3.4+ or 2.7

Install with pip

You can install from this repository using pip:

pip install git+git://github.com/BiRG/Omics-Dashboard-Python-Client.git

Usage Examples

All records can be edited in a similar fashion, but the Job record is not editable here or on the server.

Log in to service

credentials.json (stored with 400 permissions or equivalent):

{
  "email": "[email protected]",
  "password": "AVeryVeryVeryGoodPassword"
}

The Session object keeps you authenticated with the service.

from omics_dashboard_client import Session

session = Session('https://team.organization.org/omics', 'credentials.json')

You can also pass the credentials dictionary directly.

from omics_dashboard_client import Session

session = Session('https://group.institution.edu/omics', {'email': '[email protected]', 'password': 'GoodPass'})

Access collections

from omics_dashboard_client import Session, Collection
session = Session('https://example.com/omics', 'credentials.json')

# access collection #12
collection = session.get(Collection, 12)

# query all collections to get collections fitting a query
# the data received from the query is relatively small because the files are never downloaded
# search with url query parameters may be added in a later API release
collections = [collection for collection in session.get_all(Collection)
               if collection.name == 'Test Collection']

# access a collection attribute:
collection_name = collection.name

# access a collection file attribute:
proc_log = collection.get_attr('processing_log')

Modify collections

Because files may be large, they are not downloaded by default.

# modify a collection dataset
from omics_dashboard_client import Session, Collection
session = Session('https://example.com/omics', 'credentials.json')

# get a collection and modify a non-file attribute:
collection = session.get(Collection, 23)
collection.name = 'New Results'
session.update(collection)

# download the file to modify file contents:
session.download_file(collection)
Y = collection.get_dataset('Y')
Y = 0.01 * Y
collection.set_dataset('Y_modified', Y)
session.update(collection)

# create a new collection from modified collection
collection = session.get_and_download(Collection, 74)
Y = collection.get_dataset('Y')
Y = 0.01 * Y
collection.set_dataset('Y', Y)
collection.set_attr('processing_log', 'Scaled spectra by 0.01')
collection.analysis_ids = [1, 2]
session.create(collection)  # the 'id' field will change to that of the new collection, and the old one set as the parent

# Delete a record:
collection = session.get(Collection, 13)
session.delete(collection)

# Get a Pandas DataFrame containing only 'x' and 'Y'
collection = session.get(Collection, 76)
df = collection.get_dataframe(include_labels=False, numeric_columns=True)

Start a workflow on the job server

from omics_dashboard_client import Session, Workflow, Collection
session = Session('https://example.com/omics', 'credentials.json')
collection = session.get(Collection, 12)
workflow = session.get(Workflow, 7)  # Let's say the workflow does PCA on 'Y'
job_params = {
  'inputFilenames': [collection.filename]
}
job = session.submit_job(workflow, job_params)

# you can cancel the job too
msg = session.cancel_job(job)