Exemplo n.º 1
0
        splited = line.split("\t")[2:]
        if len(splited) > 5:
            msigdb_genes.append([a.strip() for a in splited])

    print(msigdb_genes)

    l = 1000
    n1 = 4
    #moj graf plus njihovi sosedi

    geneset_scores = []

    for msigdb_gene_set in msigdb_genes:
        print(k)
        k += 1
        temp = get_dict(msigdb_gene_set)  # name_officialName_dict
        if (len(temp) > 5 and len(temp) < 20):

            s = list(
                set([
                    name_interactor_dict[name] for name in list(temp)
                    if name in name_interactor_dict.keys()
                ]))
            s_with_neighbours = list(
                set([
                    item for sublist in [
                        gene_neighbours[interactor] for interactor in s
                        if interactor in gene_neighbours.keys()
                    ] for item in sublist
                ])) + s
Exemplo n.º 2
0
            unija = set(gene_neighbours_2[name_interactor_dict[temp[genes[i]]]]).union(
                set(gene_neighbours_2[name_interactor_dict[temp[genes[j]]]]))
            dm[i][j] = 1 - len(presek) / float(len(unija))

    return dm


tested = ["ECAT1", "DPPA5", "FBXO15", "NANOG", "ERAS", "DNMT3l", "ECAT8", "GDF3", "SOX15", "DPPA4", "DPPA2", "FTHL17",
              "SALL4", "REX1", "UTF1", "TCL1", "DPPA3",  "B-CATENIN",  "STAT3", "GRB2"]
simon = ["CD4", "IL2RA", "PTPRC", "FOXP3", "CTLA4", "TIGIT", "CD40LG", "THEMIS", "SATB1"]

four = ["OCT3", "OCT4", "OCT3/4", "SOX2","KLF4", "C-MYC"]

genes_with_gos = pickle.load(open('names_to_test.p', 'rb'))

temp = get_dict(tested + simon + four + genes_with_gos) # name_officialName_dict

# temp = {'OCT3':'POU5F1', 'SOX2':'SOX2', 'KLF4':'KLF4', 'C-MYC':'MYC', 'DPPA4':'DPPA4', 'FBXO15':'FBXO15', 'NANOG':'NANOG', 'ERAS':'ERAS', 'DNMT3l': 'DNMT3L', 'GDF3':'GDF3', 'SOX15':'SOX15', 'DPPA2':'DPPA2', 'SALL4':'SALL4', 'REX1':'ZFP42', 'UTF1':'UTF1', 'TCL1':'TCL1A', 'STAT3':'STAT3', 'GRB2':'GRB2', 'CD4':'CD4', 'IL2RA':'IL2RA', 'PTPRC':'PTPRC', 'FOXP3':'FOXP3', 'CTLA4':'CTLA4', 'TIGIT':'TIGIT', 'CD40LG':'CD40LG', 'THEMIS':'THEMIS', 'SATB1':'SATB1'}
# temp = {'OCT3':'POU5F1', 'SOX2':'SOX2', 'KLF4':'KLF4', 'C-MYC':'MYC', 'DPPA4':'DPPA4', 'DNMT3l': 'DNMT3L', 'UTF1':'UTF1', 'TCL1':'TCL1A', 'STAT3':'STAT3', 'GRB2':'GRB2', 'CD4':'CD4'}

plt.savefig('labels.png')

interactor_name_dict = {}
interactor_index_dict = {}

# check if Pajek NET file already exists
if exists(os.path.expanduser("~/PycharmProjects/pluripotency/Data/Homo_sapiens_interactions.net")):
#if exists( r'.\Data\Homo_sapiens_interactions.net'):
    print("tuki")
    interactor_name_dict = pickle.load( open(os.path.expanduser("~/PycharmProjects/pluripotency/Data/interactor_name_dict.p"), "rb" ) )
    interactor_index_dict = pickle.load( open(os.path.expanduser("~/PycharmProjects/pluripotency/Data/interactor_index_dict.p"), "rb" ) )
Exemplo n.º 3
0
        if len(splited) > 5:
            msigdb_genes.append([a.strip() for a in splited])

    print(msigdb_genes)

    l = 1000
    n1 = 4
#moj graf plus njihovi sosedi


    geneset_scores = []

    for msigdb_gene_set in msigdb_genes:
        print(k)
        k += 1
        temp = get_dict(msigdb_gene_set) # name_officialName_dict
        if (len(temp) > 5 and len(temp) < 20) :

            s = list(set([name_interactor_dict[name] for name in list(temp) if name in name_interactor_dict.keys()]))
            s_with_neighbours = list(set([item for sublist in [gene_neighbours[interactor] for interactor in s if interactor in gene_neighbours.keys()] for item in sublist])) + s

            S_graph = G.subgraph(s)
            S_plus_graph = G.subgraph(s_with_neighbours)

            n_S = float(S_graph.number_of_nodes())
            m_S = float(S_graph.number_of_edges())
            n = float(G.number_of_nodes())
            m = float(G.number_of_edges())

            c_S = 0
            for (u,v) in S_plus_graph.edges():