Metadata-Version: 2.0
Name: bdss-client
Version: 1.0.1b1
Summary: Big Data Smart Socket client
Home-page: https://github.com/feltus/BDSS
Author: Nick Watts
Author-email: nick@nawatts.com
License: GPLv2
Platform: UNKNOWN
Classifier: Development Status :: 4 - Beta
Classifier: Environment :: Console
Classifier: Intended Audience :: Science/Research
Classifier: License :: OSI Approved :: GNU General Public License v2 (GPLv2)
Classifier: Natural Language :: English
Classifier: Programming Language :: Python :: 3.4
Classifier: Programming Language :: Python :: 3.5
Classifier: Topic :: Scientific/Engineering
Requires-Dist: colorlog (==2.7.0)
Requires-Dist: requests (==2.8.1)
Requires-Dist: six (==1.10.0)
Requires-Dist: voluptuous (==0.8.11)

What is the Big Data Smart Socket (BDSS)?
=========================================

The increasing size of datasets used in scientific computing has made it
difficult or impossible for a researcher to store all their data at the
compute site they are using to process it. This has necessitated that a
data transfer step become a key consideration in experimental design.
Accordingly, scientific data repositories such as NCBI have begun to
offer services such as dedicated data transfer machines and advanced
transfer clients. Despite this, many researchers continue familiar but
suboptimal practices: using slow transfer clients like a web browser or
``scp``, transferring data over wireless networks, etc.

BDSS aims to alleviate this problem by shifting the burden of learning
about alternative file mirrors, transfer clients, tuning parameters,
etc. from the end user researcher to a group of "data curators". It
consists of three parts:

Components
----------

-  Metadata repository
-  Central database managed by data curators
-  Matches patterns of data file URLs and maps them to alternate sources
-  Includes information about the transfer tool to use to retrieve the
   data

-  BDSS transfer client
-  Consumes information from metadata repository
-  Invokes transfer tools
-  Reports analytics to metadata repository

-  Integration as a `Galaxy <https://galaxyproject.org/>`__ data
   transfer tool

Get Started
-----------

-  Moving data with the BDSS client:
-  `From the command
   line <https://github.com/feltus/BDSS/blob/master/client/docs>`__
-  `From
   Galaxy <https://github.com/feltus/BDSS/blob/master/galaxy_tool/README.md>`__
-  `Managing an existing metadata
   repository <https://github.com/feltus/BDSS/blob/master/metadata_repository/docs/README.md#administrator-documentation>`__
-  Setting up a new metadata repository
-  `For development or
   testing <https://github.com/feltus/BDSS/blob/master/metadata_repository/docs/developer/DevelopmentEnvironment.md>`__
-  `For production
   use <https://github.com/feltus/BDSS/blob/master/metadata_repository/docs/Installation.md>`__

Examples
--------

All examples here require a metadata repository configured to support
them. The default metadata repository at http://bdss.bioinfo.wsu.edu/
supports these examples and the necessary configuration is also listed
here.

-  `NCBI SRA
   archive <https://github.com/feltus/BDSS#ncbi-sra-archive>`__
-  `JGI Genome
   Portal <https://github.com/feltus/BDSS#jgi-genome-portal>`__

NCBI SRA archive
~~~~~~~~~~~~~~~~

NCBI makes files available for `transfer using Aspera
Connect <http://www.ncbi.nlm.nih.gov/books/NBK242625/>`__, a tool with
"improved data transfer characteristics" vs FTP or HTTP. If ``ascp`` is
installed on your machine, BDSS can handle building the appropriate
command.

Without BDSS:

.. code:: shell

    ascp -i $HOME/.aspera/connect/etc/asperaweb_id_dsa.openssh -T anonftp@ftp.ncbi.nlm.nih.gov:/sra/sra-instant/reads/ByRun/sra/SRR/SRR039/SRR039885/SRR039885.sra ./

With BDSS:

.. code:: shell

    bdss transfer -u 'ftp://ftp.ncbi.nlm.nih.gov/sra/sra-instant/reads/ByRun/sra/SRR/SRR039/SRR039885/SRR039885.sra'

Metadata repository configuration:

.. code:: json

    {
      "data_sources": [
        {
          "description": "",
          "label": "NCBI Sequence Read Archive with FTP",
          "test_files": [],
          "transfer_mechanism": {
            "options": {},
            "type": "curl"
          },
          "transforms": [
            {
              "for_destinations": [],
              "options": {
                "new_scheme": "aspera"
              },
              "target": "NCBI Sequence Read Archive with Aspera",
              "type": "change_scheme"
            }
          ],
          "url_matchers": [
            {
              "options": {
                "pattern": "^ftp://ftp\\.ncbi\\.nlm\\.nih\\.gov/sra"
              },
              "type": "regular_expression"
            }
          ]
        },
        {
          "description": "",
          "label": "NCBI Sequence Read Archive with Aspera",
          "test_files": [],
          "transfer_mechanism": {
            "options": {
              "disable_encryption": true,
              "username": "anonftp"
            },
            "type": "aspera"
          },
          "transforms": [],
          "url_matchers": [
            {
              "options": {
                "pattern": "^aspera://ftp\\.ncbi\\.nlm\\.nih\\.gov/sra"
              },
              "type": "regular_expression"
            }
          ]
        }
      ],
      "destinations": []
    }

JGI Genome Portal
~~~~~~~~~~~~~~~~~

To download files from the `JGI Genome
Portal <http://genome.jgi.doe.gov/>`__, you must first
`authenticate <http://genome.jgi.doe.gov/help/download.jsf#api>`__. BDSS
can prompt for credentials and handle storing your session cookies.

Without BDSS:

.. code:: shell

    curl 'https://signon.jgi.doe.gov/signon/create' --data-urlencode 'login=USER_NAME' --data-urlencode 'password=USER_PASSWORD' -c cookies > /dev/null
    curl 'http://genome.jgi.doe.gov/ext-api/downloads/get-directory?organism=PhytozomeV10' -b cookies > get-directory

With BDSS:

.. code:: shell

    bdss transfer -u 'http://genome.jgi.doe.gov/ext-api/downloads/get-directory?organism=PhytozomeV10'
    JGI Genome Portal username?USER_NAME
    JGI Genome Portal password?USER_PASSWORD

Metadata repository configuration:

.. code:: json

    {
      "data_sources": [
        {
          "description": "",
          "label": "JGI Genome Portal",
          "test_files": [],
          "transfer_mechanism": {
            "options": {
              "auth_url": "https://signon.jgi.doe.gov/signon/create",
              "password_field": "password",
              "password_prompt": "JGI Genome Portal password?",
              "username_field": "login",
              "username_prompt": "JGI Genome Portal username?"
            },
            "type": "session_authenticated_curl"
          },
          "transforms": [],
          "url_matchers": [
            {
              "options": {
                "pattern": "http:\\/\\/genome\\.jgi\\.doe\\.gov\\/ext-api"
              },
              "type": "regular_expression"
            }
          ]
        }
      ],
      "destinations": []
    }


