sh-miR API

Requirements

• docker
• docker-compose

That's all you need, how cool is that? :)

If you use SELinux, please allow to create volumes in directory where sh-miR designer is cloned. Assuming you're in this directory, please execute:

    $ sudo chcon -Rvt svirt_sandbox_file_t .

Quick Start

To start server:

    $ docker-compose up

You may need to run this command by root.

After this commands server builds all its dependencies and start sh-miR RESTful API.

Usage

There are three services:

• mfold
• creating sh-miR(s) from siRNA
• creating sh-miR(s) from transcript

Each of them have three URLs:

• task creator URL - */service/DATA?optional_arg1=1&optional_arg2=2
• status URL - */service/status/TASK_ID
• result URL - */service/result/TASK_ID

Where:

• service - one of services
• DATA - data which is needed to do task
• TASK_ID - id of created task

Services

• mfold
  • creator - returns TASK_ID
    URL: */mfold/DATA
  • status - returns STATUS (ok, error, fail)
    URL: */mfold/status/TASK_ID
  • result - returns zipped results of folding
    URL: */mfold/result/TASK_ID

Where DATA is a sequence which we would like to fold.

Examples:

    $ curl -i http://127.0.0.1:8080/mfold/UUUGUAUUCGCCCUAGCGC
    $ curl -i -X GET http://127.0.0.1:8080/mfold/status/f3591b31-49d9-47da-ae78-898792db26a5
    $ curl -i -X GET http://127.0.0.1:8080/mfold/result/f3591b31-49d9-47da-ae78-898792db26a5

* from_sirna * creator - returns TASK_ID
URL: `*/from_sirna/DATA` * status - returns STATUS (ok, error, fail)
URL: `*/from_sirna/status/TASK_ID` * result - returns list of tuples (score sh-miR, backbone name, pdf)
URL: `*/from_sirna/result/TASK_ID`

Where DATA is a one siRNA strand (active) or two siRNA strands separated by space (in url "%20"). First strand is active, both are in 5-3 orientation. Pdf is a task ID of mfold. To get this file use mfold service.

Examples:

    $ curl -i http://127.0.0.1:8080/from_sirna/UUUGUAUUCGCCCUAGCGC%20CGCUAUGGCGAAUACAAACA
    $ curl -i -X GET http://127.0.0.1:8080/from_sirna/status/f3591b31-49d9-47da-ae78-898792db26a5
    $ curl -i -X GET http://127.0.0.1:8080/from_sirna/result/f3591b31-49d9-47da-ae78-898792db26a5

Then we download folded sh-miR:
$ curl -i -X GET http://127.0.0.1:8080/mfold/result/4bbee83a-0337-4efa-a018-13517390ebd1

* from_transcript * creator - returns TASK_ID
URL: `*/from_transcript/DATA?optional` * min_gc - minimal GC content (default 40) * max_gc - maximal GC content (default 60) * max_offtarget - maximal offtarget (default 10) * mirna_name - name of backbone (default 'all') * stymulators - immunostimulatory sequences `['yes', 'no', 'no_difference']` (default 'no_difference') * status - returns STATUS (ok, error, fail)
URL: `*/from_transcript/status/TASK_ID` * result - returns list of dicts with keys: sh_mir, score, pdf, sequence and bacbone
URL: `*/from_transcript/result/TASK_ID`

Where DATA is transcript name from NCBI Pdf is task ID of mfold. To get this file use mfold service.

Examples:

    $ curl -i http://127.0.0.1:8080/from_transcript/NM_001164782.1
    $ curl -i -X GET http://127.0.0.1:8080/from_transcript/status/f3591b31-49d9-47da-ae78-898792db26a5
    $ curl -i -X GET http://127.0.0.1:8080/from_transcript/result/f3591b31-49d9-47da-ae78-898792db26a5

Then we download folded sh-miR:
$ curl -i -X GET http://127.0.0.1:8080/mfold/result/4bbee83a-0337-4efa-a018-13517390ebd1

Development

Debuging

Put somewhere in the code: import rpdb; rpdb.set_trace() Connecting to rpdb: $ ./scripts/rpdb_docker

Downloading own blast database

$ ./scripts/blast_docker

Clearing caches

$ ./scripts/flush

Watching celery tasks (flower):

Go to: $DOCKER_IP:5555/