hipsci.py

import requests
import json
import pandas as pd
import os
from IPython.display import display, HTML

pd.set_option("display.max_rows",10)

base_dir = "C:/Users/Pavel/Documents/Projects/hipsci/data/"
base_url = "http://www.hipsci.org/lines/api/cellLine/"

fnames = ["rs1800547", "rs8070723", "rs1052553", "rs2583988", "rs356181"]

cell_lines_json = []
with open(os.path.join(base_dir, fnames[0]), "r") as fstream:
        for line in fstream:
            cell_line_name = line.rstrip().split("\t")[0].split(".")[0]
            response = requests.get("http://www.hipsci.org/lines/api/cellLine/" + cell_line_name)
            cell_lines_json.append(json.loads(response.text))
            if (response.status_code != requests.codes.ok):
                print("ERROR: " + cell_line_name + ", status.code is " + response.status_code)

cell_line_df = pd.io.json.json_normalize(cell_lines_json).set_index(['_id'])
cell_line_df.columns = [i.replace("_source.", "") for i in cell_line_df.columns.values]

for i in fnames:
    cell_line_df[i+".haplotype"] = ""
    cell_line_df[i+".vcf_info"] = ""
    cell_line_df[i+".vcf_additional_info"] = ""
    dd = {"0|0" : "H1|H1", "0|1" : "H1|H2", "1|0" : "H1|H2", "1|1" : "H2|H2"}

for fname in fnames:
    with open(os.path.join(base_dir, fname), "r") as fstream:
        for line in fstream:
            ll = line.rstrip().split("\t")
            cell_name = ll[0].split(".")[0]
            cell_line_df.set_value(cell_name, fname + ".haplotype", dd[ll[10][:3]]) #haplo info add.info
            cell_line_df.set_value(cell_name, fname + ".vcf_info", ll[8]) #info add.info
            cell_line_df.set_value(cell_name, fname + ".vcf_additional_info", ll[10]) #add.info
            
cell_line_df.to_csv('output.txt', sep='\t')