PPI_network&calc_ppi

 # -*- coding: utf-8 -*-
 # __author__ = 'JieYao'
 from biocluster.agent import Agent
 from biocluster.tool import Tool
 import os
 import types
 import subprocess
 from biocluster.core.exceptions import OptionError

 class PpinetworkAgent(Agent):
     """
     需要calc_ppi.py
     version 1.0
     author: JieYao
     last_modified: 2016.8.15
     """

     def __init__(self, parent):
         super(PpinerworkAgent, self).__init__(parent)
         options = [
             {"name": "ppitable", "type": "infile"},
             {"name": "cut", "type": "string", "default": "-1"}
             ]
         self.add_option(options)
         self.step.add_steps('PpinetworkAnalysis')
         self.on('start', self.step_start)
         self.on('end', self.step.end)

     def step_start(self):
         self.step.PpinetworkAnalysis.start()
         self.step.update()

     def step_end(self):
         self.step.PpinetworkAnalysis.finish()
         self.step.update()

     def check_options(self):
         """
         重写参数检查
         """
         if not self.option('ppitable').is_set():
             raise OptionError('必须提供PPI网络表')
         if not os.path.exists(self.option('ppitable')):
             raise OptionError('PPI网络表路径错误')
         ppi_list = open( self.option('ppitable'), "r").readline.strip().split("\t")
         if "combined_score" not in ppi_list:
             raise OptionError("PPI网络表缺少结合分数")
         if ("yfrom" not in ppi_list) or ("to" not in ppi_list):
             raise OptionError("PPI网络缺少相互作用蛋白信息")
         try:
             eval(self.option('cut'))
         except:
             raise OptionError("Cut参数值异常，无法转换")
         return True

     def set_resource(self):
         """
         设置所需资源
         """
         self._cpu = 2
         self._memory = ''

     def end():
         result_dir = self.add_upload_dir(self.output_dir)
         result_dir.add_repath_rules([
                 [".", "", "PPI网络分析结果输出目录"],
                 ["./protein_interaction_network_centrality.txt", "txt", "PPI网络中心系数表"],
                 ["./protein_interaction_network_clustering.txt", "txt", "PPI网络节点聚类系数表"],
                 ["./protein_interaction_network_transitivity.txt", "txt", "PPI网络传递性"],
                 ["./protein_interaction_network_by_cut.txt", "txt", "Cut值约束后的PPI网络"]
                 ["./protein_interaction_network_degree_distribution.txt", "txt", "PPI网络度分布表"],
                 ["./protein_interaction_network_node_degree.txt", "txt", "PPI网络节点度属性表"]
                 ])
         print self.get_upload_files()
         super(PpinetworkAgent, self).end()

 class PpinetworkTool(Tool):
     def __init__(self, config):
         super(PpinetworkTool, self).__init__(config)
         self._version = "1.0.1"
         self.cmd_path = self.config.SOFTWARE_DIR + "/bioinfo/rna/scripts/calc_ppi.py"
         self.ppi_table = self.option('ppitable')
         self.out_files = ['protein_interaction_network_centrality.txt', 'protein_interaction_network_clustering.txt', 'protein_interaction_network_transitivity.txt', 'protein_interaction_network_by_cut.txt', 'protein_interaction_network_degree_distribution.txt', 'protein_interaction_network_node_degree.txt']

     def run(self):
         """
         运行
         """
         super(PpinetworkTool, self).run()
         self.run_ppi_network_py()

     def run_ppi_network_py(self):
         """
         运行calc_ppi.py
         """
         real_ppi_table = self.ppi_table
         cmd = self.config.SOFTWARE_DIR + '.program/Python/bin/python'
         cmd += self.cmd_path
         cmd += " -i %s -o %s" %(real_ppi_table, self.work_dir + '.ppi_network')
         if self.option("cut").is_set:
             cmd += " -c %s" %(self.option('cut'))
         self.logger.info("开始运行calc_ppi.py")

         try:
             subprocess.check_output(cmd, shell=True)
             self.logger.info('PPI_Network计算完成')
         except subprocess.CalledProcessError:
             self.logger.info('PPI_Network计算失败')
             self.ser_error("运行calc_ppi.py失败")
         allfiles = self.get_filesname()
         for i in range(len(self.out_files)):
             self.linkfile(allfiles[i], self.out_files[i])
         self.end()

     def linkfile(self, oldfile, newname):
         """
         link文件到output文件夹
         :param oldfile 资源文件路径
         :param newname 新的文件名
         :return
         """
         newpath = os.path.join(self.output_dir, newname)
         if os.path.exists(newpath):
             os.remove(newpath)
         os.link(oldfile, newpath)

     def get_filesname(self):
         files_status = [None, None, None, None, None, None]
         for paths,d,filelist in os.walk(self.work_dir + '/ppi_network'):
             for filename in filelist:
                 name = os.path.join(paths, filename)
                 for i in range(len(self.out_files)):
                     if self.out_files[i] in name:
                         files_status[i] = name
         for i in range(len(self.out_files)):
             if not files_status[i]:
                 self.set_error('未知原因，结果文件生成出错或丢失')
         return files_status

 # -*- coding: utf-8 -*-
 # __author__ = 'JieYao'

 import os
 import argparse
 from biocluster.config import Config
 import shutil
 import networkx

 global name_list
 name_list = [""]

 def search(node_name):
     global name_list
     for i in range(len(name_list)):
         if node_name == name_list[i]:
             return i
     name_list += [node_name]
     return len(name_list)-1

 parser = argparse.ArgumentParser(description='输入蛋白质相互作用网络，输出网络信息')
 parser.add_argument('-i', "--PPI_network", help="输入的PPI网络", required = True)
 parser.add_argument('-c', "--cut", help='蛋白相互作用阈值', required = False)
 parser.add_argument('-o', "--output", help = "输出文件输出路径", required = True)
 #parser.add_argument('-top', "--top", help = "First k important interaction in graph", required = False)
 args = vars(parser.parse_args())

 inFile = args["PPI_network"]
 outFile = args["output"]
 if not args["cut"]:
     cut = -1
 else:
     cut = args["cut"]

 G = networkx.Graph()
 with open(inFile, "r") as tmp_file:
     data = tmp_file.readlines()
 for i in range(1,len(data)):
     s = data[i].rstrip().split("\t")
     if eval(s[15]) >= cut:
         G.add_edge(search(s[0]), search(s[1]), weight = eval(s[15]))

 Transitivity = networkx.transitivity(G)
 Clustering = networkx.clustering(G)
 Degree_distribution = networkx.degree_histogram(G)
 Degree_Centrality = networkx.degree_centrality(G)
 Closeness_Centrality = networkx.closeness_centrality(G)
 Betweenness_Centrality = networkx.betweenness_centrality(G)
 with open(os.path.join(args["output"], "protein_interaction_network_degree_distribution.txt"), "w") as tmp_file:
     tmp_file.write("Degree\tNode_Num\n")
     for i in range(len(Degree_distribution)):
         tmp_file.write(str(i)+"\t"+str(Degree_distribution[i]))
 with open(os.path.join(args["output"], "protein_interaction_network_by_cut.txt"), "w") as tmp_file:
     tmp_file.write("Node_Num = " + str(len(G.nodes())) + "\n")
     tmp_file.write("Edge_Num = " + str(len(G.edges())) + "\n")
     tmp_file.write("Node1_Name\tNode2_Name\tWeight\n")
     for i in G.edges():
         tmp_file.write(name_list[i[0]]+"\t"+name_list[i[1]]+"\t"+str(G[i[0]][i[1]]["weight"])+"\n")
 with open(os.path.join(args["output"], "protein_interaction_network_node_degree.txt"), "w") as tmp_file:
     tmp_file.write("Node_ID\tNode_Name\tDegree\n")
     for i in range(1,len(G)+1):
         tmp_file.write(str(i)+"\t"+name_list[i]+"\t")
         tmp_file.write(str(G.degree(i))+"\n")
 with open(os.path.join(args["output"], "protein_interaction_network_centrality.txt"), "w") as tmp_file:
     tmp_file.write("Node_ID\tNode_Name\tDegree_Centrality\t")
     tmp_file.write("Closeness_Centrality\tBetweenness_Centrality\n")
     for i in range(1,len(G)+1):
         tmp_file.write(str(i)+"\t"+name_list[i]+"\t")
         tmp_file.write(str(Degree_Centrality[i])+"\t")
         tmp_file.write(str(Closeness_Centrality[i])+"\t")
         tmp_file.write(str(Betweenness_Centrality[i])+"\n")

 with open(os.path.join(args["output"], "protein_interaction_network_clustering.txt"), "w") as tmp_file:
     tmp_file.write("Node_ID\tProtein_Name\tClustering\n")
     for i in range(1,len(G)+1):
         tmp_file.write(str(i)+"\t"+name_list[i]+"\t"+str(Clustering[i])+"\n")

 with open(os.path.join(args["output"], "protein_interaction_network_transitivity.txt"), "w") as tmp_file:
     tmp_file.write("Transitivity\n")
     tmp_file.write(str(Transitivity)+"\n")

calc_ppi