WF1.py

#### Import some needed modules
import requests
import json

#### Workflow 1

#### Set the input disease
input_disease = "DOID:9352"
num_robocop_results = 25

#### Set the base URL for the reasoner and its endpoint
XRAY_API_BASE_URL = 'https://rtx.ncats.io/api/rtx/v1'
xray_url_str = XRAY_API_BASE_URL + "/query"

ROBOCOP_API_BASE_URL = 'http://robokop.renci.org/api/'
robocop_mod3_url_str = ROBOCOP_API_BASE_URL + "wf1mod3/%s/?max_results=%d" % (input_disease, num_robocop_results)
robocop_mod3a_url_str = ROBOCOP_API_BASE_URL + "wf1mod3a/%s/?max_results=%d" % (input_disease, num_robocop_results)

annot_url_str = "https://rtx.ncats.io/devED/api/rtx/v1/response/process"


################################################################
# X-ray module 1: given a disease, find genetic conditions that share "representative phenotypes" in common

#### Create a dict of the request, specifying the query type and its parameters
request = {"query_type_id": "Q10001", "terms": {"disease": input_disease}}

#### Send the request to RTX and check the status
response_content = requests.post(xray_url_str, json=request, headers={'accept': 'application/json'})
status_code = response_content.status_code
assert status_code == 200
module1_xray_results_json = response_content.json()

################################################################
# X-ray module 2: gene-centric approach
request = {"query_type_id": "Q55", "terms": {"disease": input_disease}, "bypass_cache": "true"}

#### Send the request to RTX and check the status
response_content = requests.post(xray_url_str, json=request, headers={'accept': 'application/json'})
status_code = response_content.status_code
assert status_code == 200
module2_xray_results_json = response_content.json()


################################################################
# Gamma team module 3: agent-centric
# Mod 3 un-lettered approach
response_content = requests.get(robocop_mod3_url_str, json={}, headers={'accept': 'application/json'})
status_code = response_content.status_code
assert status_code == 200
module3_robocop_results_json = response_content.json()

# Mod3a approach
response_content = requests.get(robocop_mod3a_url_str, json={}, headers={'accept': 'application/json'})
status_code = response_content.status_code
assert status_code == 200
module3a_robocop_results_json = response_content.json()

module3_robocop_results_json['reasoner_id'] = 'Robokop'  # inject the name so the website can tell them apart
module3a_robocop_results_json['reasoner_id'] = 'Robokop'
################################################################
# Orange team module 4+5: annotation and scoring

# annotate the x-ray results
to_post = {"options": ["AnnotateDrugs", "Store", "ReturnResponseId"], "responses": [module2_xray_results_json]}
module2_xray_results_annot_json = requests.post(annot_url_str, json=to_post)

# annotate gamma
to_post = {"options": ["AnnotateDrugs", "Store", "ReturnResponseId"], "responses": [module3_robocop_results_json]}
module3_robocop_results_annot_json = requests.post(annot_url_str, json=to_post)

# annotate both
to_post = {"options": ["AnnotateDrugs", "Store", "ReturnResponseId"], "responseURIs":[module2_xray_results_json['id']],"responses": [module3_robocop_results_json]}
both = requests.post(annot_url_str, json=to_post)

# Alpha's stuff
#alpha_json = json.load(open('/home/dkoslicki/Dropbox/Repositories/RTX/code/reasoningtool/QuestionAnswering/alpha_workflow-output.json','r'))
#to_post = {"options": ["AnnotateDrugs", "Store", "ReturnResponseId"], "responseURIs":[module2_xray_results_json['id']],"responses": [module3_robocop_results_json, alpha_json]}
#both = requests.post(annot_url_str, json=to_post)

################################################################
# Visualization
# The above API call creates a website (dynamically) where the results can be viewed
print("Please visit the following website:https://rtx.ncats.io/devLM/list.html?r=%s" % both.json()['response_id'])

# will return something like the following:
# https://rtx.ncats.io/devLM/list.html?r=482