1. StarBioTrek
  2. Commit 809bb0f1d3ed1af7aa63ebda4a8c805ac1619a3b
Browse code

fix

claudiacava authored on 16/10/2017 14:36:16
Showing 45 changed files
DESCRIPTION
History View file @ 809bb0f
1 1 
new file mode 100644
... ... 
@@ -0,0 +1,41 @@
1 
+Package: StarBioTrek
2 
+Type: Package
3 
+Title: StarBioTrek
4 
+Version: 1.3.2
5 
+Date: 10-16-2017
6 
+Author: Claudia Cava,
7 
+    Isabella Castiglioni
8 
+Maintainer: Claudia Cava <claudia.cava@ibfm.cnr.it>
9 
+Depends:
10 
+    R (>= 3.3)
11 
+Imports:
12 
+    SpidermiR,
13 
+	KEGGREST,
14 
+	org.Hs.eg.db,
15 
+	AnnotationDbi,
16 
+	e1071,
17 
+	ROCR,
18 
+	grDevices,
19 
+	igraph
20 
+Description: This tool StarBioTrek presents some methodologies to measure pathway activity and cross-talk among pathways integrating also the information of network data.
21 
+License: GPL (>= 3)
22 
+biocViews: GeneRegulation,
23 
+    Network,
24 
+	Pathways,
25 
+	KEGG
26 
+Suggests:
27 
+    BiocStyle,
28 
+    knitr,
29 
+    rmarkdown,
30 
+    testthat,
31 
+	devtools,
32 
+	roxygen2,
33 
+	qgraph,
34 
+	png,
35 
+	grid
36 
+VignetteBuilder: knitr
37 
+LazyData: true
38 
+URL: https://github.com/claudiacava/StarBioTrek
39 
+BugReports: https://github.com/claudiacava/StarBioTrek/issues
40 
+RoxygenNote: 6.0.1
41 
+
NAMESPACE
History View file @ 809bb0f
0 42 
new file mode 100644
... ... 
@@ -0,0 +1,35 @@
1 
+# Generated by roxygen2: do not edit by hand
2 
+
3 
+export(GE_matrix)
4 
+export(IPPI)
5 
+export(SelectedSample)
6 
+export(average)
7 
+export(ds_score_crtlk)
8 
+export(euc_dist_crtlk)
9 
+export(getKEGGdata)
10 
+export(getNETdata)
11 
+export(list_path_net)
12 
+export(matrix_plot)
13 
+export(path_net)
14 
+export(plotting_cross_talk)
15 
+export(proc_path)
16 
+export(st_dv)
17 
+export(svm_classification)
18 
+importFrom(AnnotationDbi,as.list)
19 
+importFrom(AnnotationDbi,mappedkeys)
20 
+importFrom(KEGGREST,keggGet)
21 
+importFrom(KEGGREST,keggList)
22 
+importFrom(ROCR,performance)
23 
+importFrom(ROCR,prediction)
24 
+importFrom(SpidermiR,SpidermiRanalyze_degree_centrality)
25 
+importFrom(SpidermiR,SpidermiRdownload_net)
26 
+importFrom(SpidermiR,SpidermiRprepare_NET)
27 
+importFrom(SpidermiR,SpidermiRquery_spec_networks)
28 
+importFrom(SpidermiR,SpidermiRquery_species)
29 
+importFrom(e1071,svm)
30 
+importFrom(e1071,tune)
31 
+importFrom(grDevices,rainbow)
32 
+importFrom(igraph,get.data.frame)
33 
+importFrom(igraph,graph.data.frame)
34 
+importFrom(igraph,induced.subgraph)
35 
+importFrom(org.Hs.eg.db,org.Hs.egSYMBOL2EG)
NEWS
History View file @ 809bb0f
0 36 
new file mode 100644
... ... 
@@ -0,0 +1,6 @@
1 
+  StarBioTrek
2 
+----------------------------------------------------------------
3 
+  FIRST VERSION - FEATURES
4 
+
5 
+* getKEGGdata	Searching by KEGG data.
6 
+* getNETdata	Searching by network data.
R/StarBioTrek.r
History View file @ 809bb0f
0 7 
new file mode 100644
... ... 
@@ -0,0 +1,64 @@
1 
+#' Download data
2 
+#'
3 
+#' StarBioTrek allows you to Download data of samples from StarBioTrek
4 
+#'
5 
+#' The functions you're likely to need from \pkg{StarBioTrek} is
6 
+#' \code{path_star}
7 
+#'Otherwise refer to the vignettes to see
8 
+#' how to format the documentation.
9 
+#'
10 
+#' @docType package
11 
+#' @name StarBioTrek
12 
+NULL
13 
+
14 
+#' Pathway data from KEGG
15 
+#' @docType data
16 
+#' @keywords internal
17 
+#' @name path
18 
+#' @format A data frame with rows and  variables
19 
+NULL
20 
+
21 
+#' network data
22 
+#' @docType data
23 
+#' @keywords internal
24 
+#' @name netw
25 
+#' @format A data frame with  rows and variables
26 
+NULL
27 
+
28 
+
29 
+
30 
+
31 
+#' TCGA data
32 
+#' @docType data
33 
+#' @keywords internal
34 
+#' @name Data_CANCER_normUQ_filt
35 
+#' @format A data frame with rows and variables
36 
+NULL
37 
+
38 
+#' Score Matrix of pairwise pathway using euclidean distance
39 
+#' @docType data
40 
+#' @keywords internal
41 
+#' @name score_euc_dist
42 
+#' @format A data frame with rows and variables
43 
+NULL
44 
+
45 
+#' TCGA data with normal samples
46 
+#' @docType data
47 
+#' @keywords internal
48 
+#' @name norm
49 
+#' @format A data frame with rows and variables
50 
+NULL
51 
+
52 
+#' TCGA data with tumour samples
53 
+#' @docType data
54 
+#' @keywords internal
55 
+#' @name tumo
56 
+#' @format A data frame with rows and variables
57 
+NULL
58 
+
59 
+#' A matrix of gene expression for pathways given by the user.
60 
+#' @docType data
61 
+#' @keywords internal
62 
+#' @name list_path_plot
63 
+#' @format A data frame with rows and variables
64 
+NULL
0 65 
\ No newline at end of file
R/getdata.R
History View file @ 809bb0f
1 66 
new file mode 100644
... ... 
@@ -0,0 +1,207 @@
1 
+#' @title Get human KEGG pathway data.
2 
+#' @description getKEGGdata creates a data frame with human KEGG pathway. Columns are the pathways and rows the genes inside those pathway
3 
+#' @param KEGG_path  variable
4 
+#' @export
5 
+#' @importFrom KEGGREST keggList keggGet
6 
+#' @importFrom org.Hs.eg.db org.Hs.egSYMBOL2EG
7 
+#' @importFrom AnnotationDbi mappedkeys as.list
8 
+#' @return dataframe with human pathway data
9 
+#' @examples
10 
+#' path<-getKEGGdata(KEGG_path="Transcript")
11 
+getKEGGdata<-function(KEGG_path){
12 
+if (KEGG_path=="Carb_met") {
13 
+  mer<-select_path_carb(Carbohydrate)
14 
+  c<-proc_path(mer)
15 
+  a<-c[[2]]
16 
+}
17 
+  if (KEGG_path=="Ener_met") {
18 
+    mer<-select_path_en(Energy)
19 
+    c<-proc_path(mer)
20 
+    a<-c[[2]]
21 
+  }
22 
+  if (KEGG_path=="Lip_met") {
23 
+    mer<-select_path_lip(Lipid)
24 
+    c<-proc_path(mer)
25 
+    a<-c[[2]]
26 
+  }
27 
+  if (KEGG_path=="Amn_met") {
28 
+    mer<-select_path_amn(Aminoacid)
29 
+    c<-proc_path(mer)
30 
+    a<-c[[2]]
31 
+  }
32 
+  if (KEGG_path=="Gly_bio_met") {
33 
+    mer<-select_path_gly(Glybio_met)
34 
+    c<-proc_path(mer)
35 
+    a<-c[[2]]
36 
+  }
37 
+  if (KEGG_path=="Cof_vit_met") {
38 
+    mer<-select_path_cofa(Cofa_vita_met)
39 
+    c<-proc_path(mer)
40 
+    a<-c[[2]]
41 
+  }
42 
+  if (KEGG_path=="Transcript") {
43 
+    mer<-select_path_transc(Transcription)
44 
+    c<-proc_path(mer)
45 
+    a<-c[[2]]
46 
+  }
47 
+  if (KEGG_path=="Transl") {
48 
+    mer<-select_path_transl(Translation)
49 
+    c<-proc_path(mer)
50 
+    a<-c[[2]]
51 
+  }
52 
+  if (KEGG_path=="Fold_degr") {
53 
+    mer<-select_path_fold(Folding_sorting_and_degradation)
54 
+    c<-proc_path(mer)
55 
+    a<-c[[2]]
56 
+  }
57 
+  if (KEGG_path=="Repl_repair") {
58 
+    mer<-select_path_repl(Replication_and_repair)
59 
+    c<-proc_path(mer)
60 
+    a<-c[[2]]
61 
+  }
62 
+  if (KEGG_path=="sign_transd") {
63 
+    mer<-select_path_sign(Signal_transduction)
64 
+    c<-proc_path(mer)
65 
+    a<-c[[2]]
66 
+  }
67 
+  if (KEGG_path=="sign_mol_int") {
68 
+    mer<-select_path_sign_mol(Signaling_molecules_and_interaction)
69 
+    c<-proc_path(mer)
70 
+    a<-c[[2]]
71 
+  }
72 
+  if (KEGG_path=="Transp_cat") {
73 
+    mer<-select_path_transp_ca(Transport_and_catabolism)
74 
+    c<-proc_path(mer)
75 
+    a<-c[[2]]
76 
+  }
77 
+  if (KEGG_path=="cell_grow_d") {
78 
+    mer<-select_path_cell_grow(Cell_growth_and_death)
79 
+    c<-proc_path(mer)
80 
+    a<-c[[2]]
81 
+  }
82 
+  if (KEGG_path=="cell_comm") {
83 
+    mer<-select_path_cell_comm(Cellular_community)
84 
+    c<-proc_path(mer)
85 
+    a<-c[[2]]
86 
+  }
87 
+  if (KEGG_path=="imm_syst") {
88 
+    mer<-select_path_imm_syst(Immune_system)
89 
+    c<-proc_path(mer)
90 
+    a<-c[[2]]
91 
+  }
92 
+  if (KEGG_path=="end_syst") {
93 
+    mer<-select_path_end_syst(Endocrine_system)
94 
+    c<-proc_path(mer)
95 
+    a<-c[[2]]
96 
+  }
97 
+  if (KEGG_path=="circ_syst") {
98 
+    mer<-select_path_circ_syst(Circulatory_system)
99 
+    c<-proc_path(mer)
100 
+    a<-c[[2]]
101 
+  }
102 
+  if (KEGG_path=="dig_syst") {
103 
+    mer<-select_path_dig_syst(Digestive_system)
104 
+    c<-proc_path(mer)
105 
+    a<-c[[2]]
106 
+  }
107 
+  if (KEGG_path=="exc_syst") {
108 
+    mer<-select_path_exc_syst(Excretory_system)
109 
+    c<-proc_path(mer)
110 
+    a<-c[[2]]
111 
+  }
112 
+  if (KEGG_path=="nerv_syst") {
113 
+    mer<-select_path_ner_syst(Nervous_system)
114 
+    c<-proc_path(mer)
115 
+    a<-c[[2]]
116 
+  }
117 
+  if (KEGG_path=="sens_syst") {
118 
+    mer<-select_path_sens_syst(Sensory_system)
119 
+    c<-proc_path(mer)
120 
+    a<-c[[2]]
121 
+  }
122 
+if (KEGG_path=="KEGG_path") {
123 
+  pathways.list <- keggList("pathway", "hsa")## returns the list of human pathways
124 
+pathway.codes <- sub("path:", "", names(pathways.list))
125 
+pathways.list<-list(pathways.list)
126 
+pathways.list<-pathways.list[lapply(pathways.list,length)!=0]
127 
+list_pathkeg<-do.call("cbind", pathways.list)
128 
+c<-list(pathway.codes,list_pathkeg)
129 
+a<-c[[2]]
130 
+
131 
+}
132 
+pathway.codes<-c[[1]]
133 
+genes.by.pathway <- sapply(pathway.codes,
134 
+                           function(pwid){
135 
+                             pw <- keggGet(pwid)
136 
+                             pw[[1]]$GENE[c(TRUE, FALSE)]
137 
+                           })
138 
+x <- org.Hs.egSYMBOL2EG
139 
+mapped_genes <- mappedkeys(x)
140 
+xx <- as.list(x[mapped_genes])
141 
+top3 <- matrix(0, length(xx), length(genes.by.pathway))
142 
+rownames(top3) <- names(xx)
143 
+colnames(top3)<- names(genes.by.pathway)
144 
+for (j in  1:length(xx)){
145 
+  for (k in  1:length(genes.by.pathway)){
146 
+    if (length(intersect(xx[[j]],genes.by.pathway[[k]])!=0)){
147 
+
148 
+      top3[j,k]<-names(xx[j])
149 
+    }
150 
+  }
151 
+}
152 
+top3[top3 == 0] <- " "
153 
+#a<-data.frame(pathways.list)
154 
+#i <- sapply(a, is.factor)
155 
+#a[i] <- lapply(a[i], as.character)
156 
+rownames(a)<-sub("path:","",rownames(a))
157 
+PROVA<-top3
158 
+for( i in 1:ncol(PROVA)) {
159 
+  if (colnames(PROVA)[i]==rownames(a)[i]){
160 
+    colnames(PROVA)[i]<-a[i]
161 
+}
162 
+}
163 
+return(PROVA)
164 
+}
165 
+
166 
+
167 
+#' @title Get network data.
168 
+#' @description getNETdata creates a data frame with network data.
169 
+#' Network category can be filtered among: physical interactions, co-localization, genetic interactions and shared protein domain.
170 
+#' @param network  variable. The user can use the following parameters
171 
+#' based on the network types to be used. PHint for Physical_interactions,
172 
+#' COloc for Co-localization, GENint for Genetic_interactions and
173 
+#' SHpd for Shared_protein_domains
174 
+#' @param organism organism==NULL default value is homo sapiens
175 
+#' @export
176 
+#' @importFrom SpidermiR SpidermiRquery_species SpidermiRquery_spec_networks SpidermiRdownload_net SpidermiRprepare_NET
177 
+#' @return dataframe with gene-gene (or protein-protein interactions)
178 
+#' @examples
179 
+#' organism="Saccharomyces_cerevisiae"
180 
+#' netw<-getNETdata(network="SHpd",organism)
181 
+getNETdata<-function(network,organism=NULL){
182 
+  org_shar_pro<-SpidermiRquery_species(species)
183 
+  if (is.null(organism)) {
184 
+  net_shar_prot<-SpidermiRquery_spec_networks(organismID = org_shar_pro[6,],network)
185 
+  out_net_shar_pro<-SpidermiRdownload_net(net_shar_prot)
186 
+  geneSymb_net_shar_pro<-SpidermiRprepare_NET(organismID = org_shar_pro[6,],data = out_net_shar_pro)
187 
+  }
188 
+  if( !is.null(organism) ){
189 
+    net_shar_prot<-SpidermiRquery_spec_networks(organismID = org_shar_pro[9,],network)
190 
+    out_net_shar_pro<-SpidermiRdownload_net(net_shar_prot)
191 
+    geneSymb_net_shar_pro<-SpidermiRprepare_NET(organismID = org_shar_pro[9,],data = out_net_shar_pro)
192 
+}
193 
+  ds_shar_pro<-do.call("rbind", geneSymb_net_shar_pro)
194 
+  data_shar_pro<-as.data.frame(ds_shar_pro[!duplicated(ds_shar_pro), ])
195 
+  sdc_shar_pro<-unlist(data_shar_pro$gene_symbolA,data_shar_pro$gene_symbolB)
196 
+  m_shar_pro<-c(data_shar_pro$gene_symbolA)
197 
+  m2_shar_pro<-c(data_shar_pro$gene_symbolB)
198 
+  ss_shar_pro<-cbind(m_shar_pro,m2_shar_pro)
199 
+  data_pr_shar_pro<-as.data.frame(ss_shar_pro[!duplicated(ss_shar_pro), ])
200 
+  colnames(data_pr_shar_pro) <- c("m_shar_pro", "m2_shar_pro")
201 
+return(data_pr_shar_pro)
202 
+}
203 
+
204 
+
205 
+
206 
+
207 
+
R/internal.R
History View file @ 809bb0f
0 208 
new file mode 100644
... ... 
@@ -0,0 +1,541 @@
1 
+
2 
+
3 
+
4 
+select_path_carb<-function(Carbohydrate){
5 
+species<-c("- Homo sapiens (human)")
6 
+a<-paste("Glycolysis / Gluconeogenesis", species)
7 
+b<-paste("Citrate cycle (TCA cycle)", species)
8 
+c<-paste("Pentose phosphate pathway", species)
9 
+d<-paste("Pentose and glucuronate interconversions", species)
10 
+e<-paste("Fructose and mannose metabolism", species)
11 
+f<-paste("Galactose metabolism", species)
12 
+g<-paste("Ascorbate and aldarate metabolism", species)
13 
+h<-paste("Starch and sucrose metabolism", species)
14 
+i<-paste("Amino sugar and nucleotide sugar metabolism", species)
15 
+l<-paste("Pyruvate metabolism", species)
16 
+m<-paste("Glyoxylate and dicarboxylate metabolism", species)
17 
+n<-paste("Propanoate metabolism", species)
18 
+o<-paste("Butanoate metabolism", species)
19 
+p<-paste("C5-Branched dibasic acid metabolism", species)
20 
+q<-paste("Inositol phosphate metabolism", species)
21 
+r<-paste("Enzymes", species)
22 
+s<-paste("Compounds with biological roles",species)
23 
+mer<-c(a,b,c,d,e,f,g,h,i,l,m,n,o,p,q,r,s)
24 
+return(mer)
25 
+}
26 
+
27 
+select_path_en<-function(Energy){
28 
+  species<-c("- Homo sapiens (human)")
29 
+  r<-paste("Oxidative phosphorylation", species)
30 
+  s<-paste("Photosynthesis", species)
31 
+  t<-paste("Photosynthesis - antenna proteins", species)
32 
+  v<-paste("Carbon fixation in photosynthetic organisms", species)
33 
+  u<-paste("Carbon fixation pathways in prokaryotes", species)
34 
+  z<-paste("Methane metabolism", species)
35 
+  aa<-paste("Nitrogen metabolism", species)
36 
+  ab<-paste("Sulfur metabolism", species)
37 
+  mer<-c(r,s,t,v,u,z,aa,ab)
38 
+  return(mer)
39 
+}
40 
+
41 
+
42 
+select_path_lip<-function(Lipid){
43 
+  species<-c("- Homo sapiens (human)")
44 
+ac<-paste("Fatty acid biosynthesis", species)
45 
+ad<-paste("Fatty acid elongation", species)
46 
+ae<-paste("Fatty acid degradation", species)
47 
+af<-paste("Synthesis and degradation of ketone bodies", species)
48 
+ag<-paste("Cutin, suberine and wax biosynthesis", species)
49 
+ah<-paste("Steroid biosynthesis", species)
50 
+ai<-paste("Primary bile acid biosynthesis", species)
51 
+al<-paste("Secondary bile acid biosynthesis", species)
52 
+am<-paste("Steroid hormone biosynthesis", species)
53 
+an<-paste("Glycerolipid metabolism", species)
54 
+ao<-paste("Glycerophospholipid metabolism", species)
55 
+ap<-paste("Ether lipid metabolism", species)
56 
+aq<-paste("Sphingolipid metabolism", species)
57 
+ar<-paste("Arachidonic acid metabolism", species)
58 
+as<-paste("Linoleic acid metabolism", species)
59 
+at<-paste("alpha-Linolenic acid metabolism", species)
60 
+av<-paste("Biosynthesis of unsaturated fatty acids", species)
61 
+
62 
+mer<-c(ac,ad,ae,af,ag,ah,ai,al,am,an,ao,ap,aq,ar,as,at,av)
63 
+return(mer)
64 
+}
65 
+
66 
+
67 
+
68 
+
69 
+select_path_amn<-function(Aminoacid){
70 
+  species<-c("- Homo sapiens (human)")
71 
+ac<-paste("Alanine, aspartate and glutamate metabolism", species)
72 
+ad<-paste("Glycine, serine and threonine metabolism", species)
73 
+ae<-paste("Cysteine and methionine metabolism", species)
74 
+af<-paste("Valine, leucine and isoleucine degradation", species)
75 
+ag<-paste("Valine, leucine and isoleucine biosynthesis", species)
76 
+ah<-paste("Lysine biosynthesis", species)
77 
+ai<-paste("Lysine degradation", species)
78 
+al<-paste("Arginine biosynthesis", species)
79 
+am<-paste("Arginine and proline metabolism", species)
80 
+an<-paste("Histidine metabolism", species)
81 
+ao<-paste("Tyrosine metabolism", species)
82 
+ap<-paste("Phenylalanine metabolism", species)
83 
+aq<-paste("Tryptophan metabolism", species)
84 
+ar<-paste("Phenylalanine, tyrosine and tryptophan biosynthesis", species)
85 
+as<-paste("beta-Alanine metabolism", species)
86 
+at<-paste("Taurine and hypotaurine metabolism", species)
87 
+av<-paste("Phosphonate and phosphinate metabolism", species)
88 
+au<-paste("Selenocompound metabolism", species)
89 
+az<-paste("Cyanoamino acid metabolism", species)
90 
+a<-paste("D-Glutamine and D-glutamate metabolism", species)
91 
+b<-paste("D-Arginine and D-ornithine metabolism", species)
92 
+c<-paste("D-Alanine metabolism", species)
93 
+d<-paste("Glutathione metabolism", species)
94 
+
95 
+mer<-c(ac,ad,ae,af,ag,ah,ai,al,am,an,ao,ap,aq,ar,as,at,av,au,az,a,b,c,d)
96 
+return(mer)
97 
+}
98 
+
99 
+select_path_gly<-function(Glybio_met){
100 
+  species<-c("- Homo sapiens (human)")
101 
+ac<-paste("N-Glycan biosynthesis", species)
102 
+ad<-paste("Various types of N-glycan biosynthesis", species)
103 
+ae<-paste("Mucin type O-Glycan biosynthesis", species)
104 
+af<-paste("Other types of O-glycan biosynthesis", species)
105 
+ag<-paste("Glycosaminoglycan biosynthesis - CS/DS", species)
106 
+ah<-paste("Glycosaminoglycan biosynthesis - HS/Hep", species)
107 
+ai<-paste("Glycosaminoglycan biosynthesis - KS", species)
108 
+al<-paste("Glycosaminoglycan degradation", species)
109 
+am<-paste("Glycosylphosphatidylinositol(GPI)-anchor biosynthesis", species)
110 
+an<-paste("Glycosphingolipid biosynthesis - lacto and neolacto series", species)
111 
+ao<-paste("Glycosphingolipid biosynthesis - globo series", species)
112 
+ap<-paste("Glycosphingolipid biosynthesis - ganglio series", species)
113 
+aq<-paste("Lipopolysaccharide biosynthesis", species)
114 
+ar<-paste("Peptidoglycan biosynthesis", species)
115 
+as<-paste("Other glycan degradation", species)
116 
+mer<-c(ac,ad,ae,af,ag,ah,ai,al,am,an,ao,ap,aq,ar,as)
117 
+return(mer)
118 
+}
119 
+
120 
+
121 
+
122 
+select_path_cofa<-function(Cofa_vita_met){
123 
+  species<-c("- Homo sapiens (human)")
124 
+ac<-paste("Thiamine metabolism", species)
125 
+ad<-paste("Riboflavin metabolism", species)
126 
+ae<-paste("Vitamin B6 metabolism", species)
127 
+af<-paste("Nicotinate and nicotinamide metabolism", species)
128 
+ag<-paste("Pantothenate and CoA biosynthesis", species)
129 
+ah<-paste("Biotin metabolism", species)
130 
+ai<-paste("Lipoic acid metabolism", species)
131 
+al<-paste("Folate biosynthesis", species)
132 
+am<-paste("One carbon pool by folate", species)
133 
+an<-paste("Retinol metabolism", species)
134 
+ao<-paste("Porphyrin and chlorophyll metabolism", species)
135 
+ap<-paste("Ubiquinone and other terpenoid-quinone biosynthesis", species)
136 
+mer<-c(ac,ad,ae,af,ag,ah,ai,al,am,an,ao,ap)
137 
+return(mer)
138 
+}
139 
+
140 
+select_path_transc<-function(Transcription){
141 
+  species<-c("- Homo sapiens (human)")
142 
+ac<-paste("RNA polymerase", species)
143 
+ad<-paste("Basal transcription factors", species)
144 
+ae<-paste("Spliceosome", species)
145 
+af<-paste("Transcription factors", species)
146 
+ag<-paste("Transcription machinery", species)
147 
+mer<-c(ac,ad,ae,af,ag)
148 
+return(mer)
149 
+}
150 
+
151 
+
152 
+
153 
+select_path_transl<-function(Translation){
154 
+  species<-c("- Homo sapiens (human)")
155 
+ac<-paste("Ribosome", species)
156 
+ad<-paste("Aminoacyl-tRNA biosynthesis", species)
157 
+ae<-paste("RNA transport", species)
158 
+af<-paste("mRNA surveillance pathway", species)
159 
+ag<-paste("Ribosome biogenesis in eukaryotes", species)
160 
+ah<-paste("Ribosomal proteins", species)
161 
+ai<-paste("Ribosome biogenesis", species)
162 
+al<-paste("Transfer RNA biogenesis", species)
163 
+am<-paste("Translation factors", species)
164 
+mer<-c(ac,ad,ae,af,ag,ah,ai,al,am)
165 
+return(mer)
166 
+}
167 
+
168 
+select_path_fold<-function(Folding_sorting_and_degradation){
169 
+  species<-c("- Homo sapiens (human)")
170 
+ac<-paste("Protein export", species)
171 
+ad<-paste("Protein processing in endoplasmic reticulum", species)
172 
+ae<-paste("SNARE interactions in vesicular transport", species)
173 
+af<-paste("Ubiquitin mediated proteolysis", species)
174 
+ag<-paste("Sulfur relay system", species)
175 
+ah<-paste("RNA degradation", species)
176 
+ai<-paste("Chaperones and folding catalysts", species)
177 
+al<-paste("SNAREs", species)
178 
+am<-paste("Ubiquitin system", species)
179 
+an<-paste("Proteasome", species)
180 
+mer<-c(ac,ad,ae,af,ag,ah,ai,al,am,an)
181 
+return(mer)
182 
+}
183 
+
184 
+
185 
+
186 
+
187 
+select_path_repl<-function(Replication_and_repair){
188 
+  species<-c("- Homo sapiens (human)")
189 
+ac<-paste("DNA replication", species)
190 
+ad<-paste("Base excision repair", species)
191 
+ae<-paste("Nucleotide excision repair", species)
192 
+af<-paste("Mismatch repair", species)
193 
+ag<-paste("Homologous recombination", species)
194 
+ah<-paste("Non-homologous end-joining", species)
195 
+ai<-paste("Fanconi anemia pathway", species)
196 
+al<-paste("DNA replication proteins", species)
197 
+am<-paste("Chromosome", species)
198 
+an<-paste("DNA repair and recombination", species)
199 
+ao<-paste("proteins", species)
200 
+mer<-c(ac,ad,ae,af,ag,ah,ai,al,am,an,ao)
201 
+return(mer)
202 
+}
203 
+
204 
+
205 
+
206 
+select_path_sign<-function(Signal_transduction){
207 
+  species<-c("- Homo sapiens (human)")
208 
+a<-paste("Ras signaling pathway", species)
209 
+b<-paste("Rap1 signaling pathway", species)
210 
+c<-paste("MAPK signaling pathway", species)
211 
+d<-paste("ErbB signaling pathway", species)
212 
+e<-paste("Wnt signaling pathway", species)
213 
+f<-paste("Notch signaling pathway", species)
214 
+g<-paste("Hedgehog signaling pathway", species)
215 
+h<-paste("TGF-beta signaling pathway", species)
216 
+i<-paste("Hippo signaling pathway", species)
217 
+l<-paste("VEGF signaling pathway", species)
218 
+m<-paste("Jak-STAT signaling pathway", species)
219 
+n<-paste("NF-kappa B signaling pathway", species)
220 
+o<-paste("TNF signaling pathway", species)
221 
+p<-paste("HIF-1 signaling pathway", species)
222 
+q<-paste("FoxO signaling pathway", species)
223 
+r<-paste("Calcium signaling pathway", species)
224 
+s<-paste("Phosphatidylinositol signaling system", species)
225 
+t<-paste("Phospholipase D signaling pathway", species)
226 
+v<-paste("Sphingolipid signaling pathway", species)
227 
+u<-paste("cAMP signaling pathway", species)
228 
+z<-paste("cGMP-PKG signaling pathway", species)
229 
+ab<-paste("PI3K-Akt signaling pathway", species)
230 
+ac<-paste("AMPK signaling pathway", species)
231 
+ad<-paste("mTOR signaling pathway", species)
232 
+mer<-c(a,b,c,d,e,f,g,h,i,l,m,n,o,p,q,r,s,t,v,u,z,ab,ac,ad)
233 
+return(mer)
234 
+}
235 
+
236 
+
237 
+select_path_sign_mol<-function(Signaling_molecules_and_interaction){
238 
+  species<-c("- Homo sapiens (human)")
239 
+a<-paste("Neuroactive ligand-receptor interaction", species)
240 
+b<-paste("Cytokine-cytokine receptor interaction", species)
241 
+c<-paste("ECM-receptor interaction", species)
242 
+d<-paste("Cell adhesion molecules (CAMs)", species)
243 
+mer<-c(a,b,c,d)
244 
+return(mer)
245 
+}
246 
+
247 
+
248 
+select_path_transp_ca<-function(Transport_and_catabolism){
249 
+  species<-c("- Homo sapiens (human)")
250 
+a<-paste("Endocytosis", species)
251 
+b<-paste("Phagosome", species)
252 
+c<-paste("Lysosome", species)
253 
+d<-paste("Peroxisome", species)
254 
+e<-paste("Regulation of autophagy", species)
255 
+mer<-c(a,b,c,d,e)
256 
+return(mer)
257 
+}
258 
+
259 
+select_path_cell_grow<-function(Cell_growth_and_death){
260 
+  species<-c("- Homo sapiens (human)")
261 
+  a<-paste("Cell cycle", species)
262 
+b<-paste("Apoptosis", species)
263 
+c<-paste("p53 signaling pathway", species)
264 
+mer<-c(a,b,c)
265 
+return(mer)
266 
+}
267 
+
268 
+
269 
+select_path_cell_comm<-function(Cellular_community){
270 
+  species<-c("- Homo sapiens (human)")
271 
+  a<-paste("Focal adhesion", species)
272 
+b<-paste("Adherens junction", species)
273 
+c<-paste("Tight junction", species)
274 
+d<-paste("Gap junction", species)
275 
+e<-paste("Signaling pathways regulating pluripotency of stem cells ", species)
276 
+mer<-c(a,b,c,d,e)
277 
+return(mer)
278 
+}
279 
+
280 
+
281 
+select_path_imm_syst<-function(Immune_system){
282 
+  species<-c("- Homo sapiens (human)")
283 
+a<-paste("Hematopoietic cell lineage", species)
284 
+b<-paste("Complement and coagulation cascades", species)
285 
+c<-paste("Platelet activation", species)
286 
+d<-paste("Toll-like receptor signaling pathway", species)
287 
+e<-paste("Toll and Imd signaling pathway", species)
288 
+f<-paste("NOD-like receptor signaling pathway", species)
289 
+g<-paste("RIG-I-like receptor signaling pathway", species)
290 
+h<-paste("Cytosolic DNA-sensing pathway", species)
291 
+i<-paste("Natural killer cell mediated cytotoxicity", species)
292 
+l<-paste("Antigen processing and presentation", species)
293 
+m<-paste("T cell receptor signaling pathway", species)
294 
+n<-paste("B cell receptor signaling pathway", species)
295 
+o<-paste("Fc epsilon RI signaling pathway", species)
296 
+p<-paste("Fc gamma R-mediated phagocytosis", species)
297 
+q<-paste("Leukocyte transendothelial migration", species)
298 
+r<-paste("Intestinal immune network for IgA production", species)
299 
+s<-paste("Chemokine signaling pathway", species)
300 
+
301 
+mer<-c(a,b,c,d,e,f,g,h,i,l,m,n,o,p,q,r,s)
302 
+return(mer)
303 
+}
304 
+
305 
+
306 
+
307 
+
308 
+select_path_end_syst<-function(Endocrine_system){
309 
+  species<-c("- Homo sapiens (human)")
310 
+a<-paste("Insulin secretion", species)
311 
+b<-paste("Insulin signaling pathway", species)
312 
+c<-paste("Glucagon signaling pathway", species)
313 
+d<-paste("Regulation of lipolysis in adipocytes", species)
314 
+e<-paste("Adipocytokine signaling pathway", species)
315 
+f<-paste("PPAR signaling pathway", species)
316 
+g<-paste("GnRH signaling pathway", species)
317 
+h<-paste("Ovarian steroidogenesis", species)
318 
+i<-paste("Estrogen signaling pathway", species)
319 
+l<-paste("Progesterone-mediated oocyte maturation", species)
320 
+m<-paste("Prolactin signaling pathway", species)
321 
+n<-paste("Oxytocin signaling pathway", species)
322 
+o<-paste("Thyroid hormone synthesis", species)
323 
+p<-paste("Thyroid hormone signaling pathway", species)
324 
+q<-paste("Melanogenesis", species)
325 
+r<-paste("Renin secretion", species)
326 
+s<-paste("Renin-angiotensin system", species)
327 
+t<-paste("Aldosterone synthesis and secretion", species)
328 
+
329 
+
330 
+mer<-c(a,b,c,d,e,f,g,h,i,l,m,n,o,p,q,r,s,t)
331 
+return(mer)
332 
+}
333 
+
334 
+
335 
+select_path_circ_syst<-function(Circulatory_system){
336 
+  species<-c("- Homo sapiens (human)")
337 
+  a<-paste("Cardiac muscle contraction", species)
338 
+b<-paste("Adrenergic signaling in cardiomyocytes", species)
339 
+c<-paste("Vascular smooth muscle contraction", species)
340 
+mer<-c(a,b,c)
341 
+return(mer)
342 
+}
343 
+
344 
+
345 
+select_path_dig_syst<-function(Digestive_system){
346 
+  species<-c("- Homo sapiens (human)")
347 
+  a<-paste("Salivary secretion", species)
348 
+b<-paste("Gastric acid secretion", species)
349 
+c<-paste("Pancreatic secretion", species)
350 
+d<-paste("Bile secretion", species)
351 
+e<-paste("Carbohydrate digestion and absorption", species)
352 
+f<-paste("Protein digestion and absorption", species)
353 
+g<-paste("Fat digestion and absorption", species)
354 
+h<-paste("Vitamin digestion and absorption", species)
355 
+i<-paste("Mineral absorption", species)
356 
+
357 
+mer<-c(a,b,c,d,e,f,g,h,i)
358 
+return(mer)
359 
+}
360 
+
361 
+
362 
+
363 
+select_path_exc_syst<-function(Excretory_system){
364 
+  species<-c("- Homo sapiens (human)")
365 
+  a<-paste("Vasopressin-regulated water reabsorption", species)
366 
+b<-paste("Aldosterone-regulated sodium reabsorption", species)
367 
+c<-paste("Endocrine and other factor-regulated calcium reabsorption", species)
368 
+d<-paste("Proximal tubule bicarbonate reclamation", species)
369 
+e<-paste("Collecting duct acid secretion", species)
370 
+
371 
+
372 
+mer<-c(a,b,c,d,e)
373 
+return(mer)
374 
+}
375 
+
376 
+
377 
+select_path_ner_syst<-function(Nervous_system){
378 
+  species<-c("- Homo sapiens (human)")
379 
+a<-paste("Glutamatergic synapse", species)
380 
+b<-paste("GABAergic synapse", species)
381 
+c<-paste("Cholinergic synapse", species)
382 
+d<-paste("Dopaminergic synapse", species)
383 
+e<-paste("Serotonergic synapse", species)
384 
+f<-paste("Long-term potentiation", species)
385 
+g<-paste("Long-term depression", species)
386 
+h<-paste("Retrograde endocannabinoid signaling", species)
387 
+i<-paste("Synaptic vesicle cycle", species)
388 
+l<-paste("Neurotrophin signaling pathway", species)
389 
+
390 
+mer<-c(a,b,c,d,e,f,g,h,i,l)
391 
+return(mer)
392 
+}
393 
+
394 
+
395 
+select_path_sens_syst<-function(Sensory_system){
396 
+  species<-c("- Homo sapiens (human)")
397 
+  a<-paste("Phototransduction", species)
398 
+b<-paste("Olfactory transduction", species)
399 
+c<-paste("Taste transduction", species)
400 
+d<-paste("Inflammatory mediator regulation of TRP channels", species)
401 
+mer<-c(a,b,c,d)
402 
+return(mer)
403 
+}
404 
+
405 
+
406 
+
407 
+#' @title Select the class of TCGA data
408 
+#' @description select two labels from ID barcode
409 
+#' @param Dataset gene expression matrix
410 
+#' @param typesample the labels of the samples (e.g. tumor,normal)
411 
+#' @export
412 
+#' @return a gene expression matrix of the samples with specified label
413 
+#' @examples
414 
+#' tumo<-SelectedSample(Dataset=Data_CANCER_normUQ_filt,typesample="tumor")[,2]
415 
+SelectedSample <- function(Dataset,typesample){
416 
+  if( typesample =="tumor"){
417 
+    Dataset <- Dataset[,which( as.numeric(substr(colnames(Dataset), 14, 15)) == 01) ]
418 
+  }
419 
+
420 
+  if( typesample =="normal"){
421 
+    Dataset <- Dataset[,which( as.numeric(substr(colnames(Dataset), 14, 15)) >= 10) ]
422 
+  }
423 
+
424 
+  return(Dataset)
425 
+
426 
+}
427 
+
428 
+
429 
+#' @title Select the class of TCGA data
430 
+#' @description select two labels from ID barcode
431 
+#' @param cutoff cut-off for AUC value
432 
+#' @param auc.df list of AUC value
433 
+#' @return a gene expression matrix with only pairwise pathway with a particular cut-off
434 
+select_class<-function(auc.df,cutoff){
435 
+ds<-do.call("rbind", auc.df)
436 
+tmp_ordered <- as.data.frame(ds[order(ds,decreasing=TRUE),])
437 
+colnames(tmp_ordered)<-'pathway'
438 
+er<-as.data.frame(tmp_ordered$pathway>cutoff)
439 
+ase<-tmp_ordered[tmp_ordered$pathway>cutoff,]
440 
+rownames(er)<-rownames(tmp_ordered)
441 
+er[,2]<-tmp_ordered$pathway
442 
+lipid_metabolism<-er[1:length(ase),]
443 
+return(lipid_metabolism)
444 
+}
445 
+
446 
+
447 
+
448 
+
449 
+#' @title Process matrix TCGA data after the selection of pairwise pathway
450 
+#' @description processing gene expression matrix
451 
+#' @param measure matrix with measure of cross-talk among pathways
452 
+#' @param list_perf output of the function select_class
453 
+#' @return a gene expression matrix for case study 1
454 
+process_matrix<-function(measure,list_perf){
455 
+scoreMatrix <- as.data.frame(measure[,3:ncol(measure)])
456 
+for( i in 1: ncol(scoreMatrix)){
457 
+  scoreMatrix[,i] <- as.numeric(as.character(scoreMatrix[,i]))
458 
+}
459 
+measure[,1] <- gsub(" ", "_", measure[,1])
460 
+d<-sub('_-_Homo_sapiens_*', '', measure[,1])
461 
+d_pr<- gsub("(human)", "", d, fixed="TRUE")
462 
+d_pr <- gsub("_", "", d_pr)
463 
+d_pr <- gsub("-", "", d_pr)
464 
+measure[,2] <- gsub(" ", "_", measure[,2])
465 
+d2<-sub('_-_Homo_sapiens_(human)*', '', measure[,2])
466 
+d_pr2<- gsub("(human)", "", d2, fixed="TRUE")
467 
+d_pr2 <- gsub("_", "", d_pr2)
468 
+d_pr2 <- gsub("-", "", d_pr2)
469 
+PathwaysPair <- paste( as.matrix(d_pr), as.matrix(d_pr2),sep="_" )
470 
+rownames(scoreMatrix) <-PathwaysPair
471 
+intera<-intersect(rownames(scoreMatrix),rownames(list_perf))
472 
+path_bestlipd<-scoreMatrix[intera,]
473 
+return(path_bestlipd)
474 
+}
475 
+
476 
+
477 
+
478 
+process_matrix_cell_process<-function(measure_cell_process){
479 
+score__cell_grow_d <- as.data.frame(measure_cell_process[,3:ncol(measure_cell_process)])
480 
+for( i in 1: ncol(score__cell_grow_d)){
481 
+  score__cell_grow_d[,i] <- as.numeric(as.character(score__cell_grow_d[,i]))
482 
+}
483 
+
484 
+measure_cell_process[,1] <- gsub(" ", "_", measure_cell_process[,1])
485 
+d<-sub('_-_Homo_sapiens_*', '', measure_cell_process[,1])
486 
+
487 
+d_pr<- gsub("(human)", "", d, fixed="TRUE")
488 
+d_pr <- gsub("_", "", d_pr)
489 
+d_pr <- gsub("-", "", d_pr)
490 
+
491 
+measure_cell_process[,2] <- gsub(" ", "_", measure_cell_process[,2])
492 
+d2<-sub('_-_Homo_sapiens_(human)*', '', measure_cell_process[,2])
493 
+d_pr2<- gsub("(human)", "", d2, fixed="TRUE")
494 
+d_pr2 <- gsub("_", "", d_pr2)
495 
+d_pr2 <- gsub("-", "", d_pr2)
496 
+
497 
+PathwaysPair <- paste( as.matrix(d_pr), as.matrix(d_pr2),sep="_" )
498 
+rownames(score__cell_grow_d) <-PathwaysPair
499 
+return(score__cell_grow_d)
500 
+}
501 
+
502 
+
503 
+#' @title Get human KEGG pathway data.
504 
+#' @description getKEGGdata creates a data frame with human KEGG pathway. Columns are the pathways and rows the genes inside those pathway
505 
+#' @param mer  output for example of select_path_carb
506 
+#' @export
507 
+#' @importFrom KEGGREST keggList
508 
+#' @return dataframe with human pathway data
509 
+proc_path<-function(mer){
510 
+pathways.list <- keggList("pathway", "hsa")## returns the list of human pathways
511 
+common<-intersect(pathways.list,mer)
512 
+lo<-list()
513 
+for (i in 1:length(pathways.list)){
514 
+  if (length(intersect(pathways.list[[i]],common)!=0)){
515 
+    lo[[i]]<-pathways.list[[i]]
516 
+    names(lo)[[i]]<-names(pathways.list)[[i]]
517 
+  }
518 
+}
519 
+pathways.list<-lo[lapply(lo,length)!=0]
520 
+pathway.codes <- sub("path:", "", names(pathways.list))
521 
+b<-do.call("rbind", pathways.list)
522 
+list_pathkegg<-list(pathway.codes,b)
523 
+return(list_pathkegg)
524 
+}
525 
+
526 
+
527 
+
528 
+delete.NULLs  <-  function(xlist){   # delele null/empty entries in a list
529 
+  xlist[unlist(lapply(xlist, nrow) != 0)]
530 
+}
531 
+
532 
+
533 
+
534 
+
535 
+
536 
+
537 
+
538 
+
539 
+
540 
+
541 
+
R/path_star.R
History View file @ 809bb0f
0 542 
new file mode 100644
... ... 
@@ -0,0 +1,570 @@
1 
+#' @title Get human KEGG pathway data and network data in order to define the common gene.
2 
+#' @description path_net creates a list of network data for each human pathway. The network data will be generated when interacting genes belong to that pathway.
3 
+#' @param data  network data as provided by getNETdata
4 
+#' @param pathway  pathway data as provided by getKEGGdata
5 
+#' @importFrom igraph graph.data.frame induced.subgraph get.data.frame
6 
+#' @export
7 
+#' @return a list of network data for each pathway (interacting genes belong to that pathway)
8 
+#' @examples
9 
+#' lista_net<-path_net(pathway=path,data=netw)
10 
+path_net<-function(pathway,data){
11 
+  lista_int<-list()
12 
+  for (k in 1:ncol(pathway)){
13 
+    print(colnames(pathway)[k])
14 
+    currentPathway_genes<-pathway[,k]
15 
+    colnames(data) <- c("gene_symbolA", "gene_symbolB")
16 
+    i <- sapply(data, is.factor)
17 
+    data[i] <- lapply(data[i], as.character)
18 
+    ver<-unlist(data)
19 
+    n<-unique(ver)
20 
+    s<-intersect(n,currentPathway_genes)
21 
+    g <- graph.data.frame(data,directed=FALSE)
22 
+    g2 <- induced.subgraph(graph=g,vids=s)
23 
+    aaa<-get.data.frame(g2)
24 
+    colnames(aaa)[1] <- 'V1'
25 
+    colnames(aaa)[2] <- 'V2'
26 
+    lista_int[[k]]<-aaa
27 
+    names(lista_int)[k]<-colnames(pathway)[k]
28 
+  }
29 
+  return(lista_int)
30 
+}
31 
+
32 
+
33 
+
34 
+
35 
+#' @title Get human KEGG pathway data and output of path_net in order to define the common genes.
36 
+#' @description list_path_net creates a list of interacting genes for each human pathway.
37 
+#' @param lista_net  output of path_net
38 
+#' @param pathway  pathway data as provided by getKEGGdata
39 
+#' @export
40 
+#' @return a list of genes for each pathway (interacting genes belong to that pathway)
41 
+#' @examples
42 
+#' lista_netw<-path_net(pathway=path,data=netw)
43 
+#' list_path<-list_path_net(lista_net=lista_netw,pathway=path)
44 
+list_path_net<-function(lista_net,pathway){
45 
+v=list()
46 
+bn=list()
47 
+for (j in 1:length(lista_net)){
48 
+  cf<-lista_net[[j]]
49 
+  i <- sapply(cf, is.factor)
50 
+  cf[i] <- lapply(cf[i], as.character)
51 
+  colnames(cf) <- c("m_shar_pro", "m2_shar_pro")
52 
+  m<-c(cf$m_shar_pro)
53 
+  m2<-c(cf$m2_shar_pro)
54 
+  s<-c(m,m2)
55 
+  fr<- unique(s)
56 
+  n<-as.data.frame(fr)
57 
+  if(length(n)==0){
58 
+    v[[j]]<-NULL
59 
+
60 
+  }
61 
+  if(length(n)!=0){
62 
+  i <- sapply(n, is.factor)
63 
+  n[i] <- lapply(n[i], as.character)
64 
+  #for (k in  1:ncol(pathway)){
65 
+  if (length(intersect(n$fr,pathway[,j]))==nrow(n)){
66 
+    print(paste("List of genes interacting in the same pathway:",colnames(pathway)[j]))
67 
+    aa<-intersect(n$fr,pathway[,j])
68 
+    v[[j]]<-aa
69 
+    names(v)[j]<-colnames(pathway)[j]
70 
+  }
71 
+}}
72 
+return(v)}
73 
+
74 
+
75 
+
76 
+
77 
+#' @title Get human KEGG pathway data and a gene expression matrix in order to obtain a matrix with the gene expression for only pathways given in input .
78 
+#' @description GE_matrix creates a matrix of gene expression for pathways given by the user.
79 
+#' @param DataMatrix  gene expression matrix (eg.TCGA data)
80 
+#' @param pathway  pathway data as provided by getKEGGdata
81 
+#' @export
82 
+#' @return a matrix for each pathway ( gene expression level belong to that pathway)
83 
+#' @examples
84 
+#' list_path_gene<-GE_matrix(DataMatrix=tumo[,1:2],pathway=path)
85 
+GE_matrix<-function(DataMatrix,pathway) {
86 
+  path_name<-sub(' ', '_',colnames(pathway))
87 
+d_pr<- gsub(" - Homo sapiens (human)", "", path_name, fixed="TRUE")
88 
+colnames(pathway)<-d_pr
89 
+#zz<-as.data.frame(rowMeans(DataMatrix))
90 
+zz<-as.data.frame(DataMatrix)
91 
+v<-list()
92 
+for ( k in 1: ncol(pathway)){
93 
+  #k=2
94 
+  if (length(intersect(rownames(zz),pathway[,k])!=0)){
95 
+    print(colnames(path)[k])
96 
+  currentPathway_genes_list_common <- intersect(rownames(zz), currentPathway_genes<-pathway[,k])
97 
+  currentPathway_genes_list_commonMatrix <- as.data.frame(zz[currentPathway_genes_list_common,])
98 
+  rownames(currentPathway_genes_list_commonMatrix)<-currentPathway_genes_list_common
99 
+  v[[k]]<- currentPathway_genes_list_commonMatrix
100 
+  names(v)[k]<-colnames(pathway)[k]
101 
+  }
102 
+}
103 
+#PEAmatrix <- matrix( 0,nrow(DataMatrix),ncol(pathway))
104 
+#rownames(PEAmatrix) <- as.factor(rownames(DataMatrix))
105 
+#colnames(PEAmatrix) <-  as.factor(colnames(pathway))
106 
+#for (i in 1:length(v)){
107 
+#PEAmatrix[v[[i]],i]<-zz[v[[i]],]
108 
+#}
109 
+#PEAmatrix<-PEAmatrix[which(rowSums(PEAmatrix) > 0),]
110 
+return(v)
111 
+}
112 
+
113 
+
114 
+
115 
+#' @title Get human KEGG pathway data and a gene expression matrix in order to obtain a matrix with the mean gene expression for only pathways given in input .
116 
+#' @description GE_matrix creates a matrix of mean gene expression for pathways given by the user.
117 
+#' @param DataMatrix  gene expression matrix (eg.TCGA data)
118 
+#' @param pathway  pathway data as provided by getKEGGdata
119 
+#' @export
120 
+#' @return a matrix for each pathway (mean gene expression level belong to that pathway)
121 
+#' @examples
122 
+#' list_path_plot<-matrix_plot(DataMatrix=tumo[,1:2],pathway=path)
123 
+matrix_plot<-function(DataMatrix,pathway) {
124 
+  path_name<-sub(' ', '_',colnames(pathway))
125 
+  d_pr<- gsub(" - Homo sapiens (human)", "", path_name, fixed="TRUE")
126 
+  colnames(pathway)<-d_pr
127 
+  zz<-as.data.frame(rowMeans(DataMatrix))
128 
+  v<-list()
129 
+  for ( k in 1: ncol(pathway)){
130 
+    #k=2
131 
+    if (length(intersect(rownames(zz),pathway[,k])!=0)){
132 
+      print(colnames(path)[k])
133 
+      currentPathway_genes_list_common <- intersect(rownames(zz), currentPathway_genes<-pathway[,k])
134 
+      currentPathway_genes_list_commonMatrix <- as.data.frame(zz[currentPathway_genes_list_common,])
135 
+      rownames(currentPathway_genes_list_commonMatrix)<-currentPathway_genes_list_common
136 
+      v[[k]]<- currentPathway_genes_list_common
137 
+      names(v)[k]<-colnames(pathway)[k]
138 
+    }
139 
+  }
140 
+  PEAmatrix <- matrix( 0,nrow(DataMatrix),ncol(pathway))
141 
+  rownames(PEAmatrix) <- as.factor(rownames(DataMatrix))
142 
+  colnames(PEAmatrix) <-  as.factor(colnames(pathway))
143 
+  for (i in 1:length(v)){
144 
+  PEAmatrix[v[[i]],i]<-zz[v[[i]],]
145 
+  }
146 
+  PEAmatrix<-PEAmatrix[which(rowSums(PEAmatrix) > 0),]
147 
+  return(PEAmatrix)
148 
+}
149 
+
150 
+
151 
+
152 
+
153 
+
154 
+
155 
+
156 
+
157 
+
158 
+
159 
+
160 
+
161 
+
162 
+#' @title Get human KEGG pathway data and a gene expression matrix we obtain a matrix with the gene expression for only pathways given in input .
163 
+#' @description plotting_matrix creates a matrix of gene expression for pathways given by the user.
164 
+#' @param DataMatrix  gene expression matrix (eg.TCGA data)
165 
+#' @param pathway  pathway data as provided by getKEGGdata
166 
+#' @param path_matrix  output of the function matrix_plot
167 
+#' @export
168 
+#' @return a plot for pathway cross talk
169 
+#' @examples
170 
+#' mt<-plotting_cross_talk(DataMatrix=tumo[,1:2],pathway=path,path_matrix=list_path_plot)
171 
+plotting_cross_talk<-function(DataMatrix,pathway,path_matrix){
172 
+  zz<-as.data.frame(rowMeans(DataMatrix))
173 
+  v<-list()
174 
+  for ( k in 1: ncol(pathway)){
175 
+    path_name<-sub(' ', '_',colnames(pathway))
176 
+    d_pr<- gsub(" - Homo sapiens (human)", "", path_name, fixed="TRUE")
177 
+    colnames(pathway)<-d_pr
178 
+    if (length(intersect(rownames(zz),pathway[,k])!=0)){
179 
+      print(colnames(path)[k])
180 
+      currentPathway_genes_list_common <- intersect(rownames(zz), currentPathway_genes<-pathway[,k])
181 
+      currentPathway_genes_list_commonMatrix <- as.data.frame(zz[currentPathway_genes_list_common,])
182 
+      rownames(currentPathway_genes_list_commonMatrix)<-currentPathway_genes_list_common
183 
+      v[[k]]<- as.factor(currentPathway_genes_list_common)
184 
+      names(v)[k]<-colnames(pathway)[k]
185 
+    }
186 
+  }
187 
+  vv<-list()
188 
+  mi<-t(path_matrix)
189 
+
190 
+  dc<-cor(mi)
191 
+  for ( k in 1: length(v)){
192 
+    currentPathway_genes_list_common <- intersect(rownames(dc), v[[k]])
193 
+    a<-match(currentPathway_genes_list_common,rownames(dc))
194 
+    vv[[k]]<- a
195 
+    names(vv)[k]<-colnames(pathway)[k]
196 
+  }
197 
+  list_plt=list(corr=dc,gruppi=vv)
198 
+ #r<-qgraph(list_plt$corr, groups=list_plt$gruppi, mar=c(1,1,1,1),minimum=0.6)
199 
+  return(list_plt)
200 
+}
201 
+
202 
+
203 
+
204 
+
205 
+#' @title For TCGA data get human pathway data and creates a matrix with the average of genes for each pathway.
206 
+#' @description average creates a matrix with a summarized value for each pathway
207 
+#' @param dataFilt TCGA matrix
208 
+#' @param pathway pathway data
209 
+#' @export
210 
+#' @return a matrix value for each pathway
211 
+#' @examples
212 
+#' score_mean<-average(dataFilt=tumo[,1:2],path)
213 
+average<-function(dataFilt,pathway){
214 
+  DataMatrix<-dataFilt
215 
+  #dataFilt[ , "new.col"] <- gsub("\\|.*", "", rownames(dataFilt))
216 
+  #DataMatrix<-dataFilt[which(dataFilt$new.col!="?"),]
217 
+  #DataMatrix <- subset(DataMatrix, !duplicated(DataMatrix$new.col))
218 
+  #rownames(DataMatrix)<-DataMatrix$new.col
219 
+  #DataMatrix$new.col<-NULL
220 
+
221 
+PEAmatrix <- matrix( 0, ncol(pathway),ncol(DataMatrix))
222 
+rownames(PEAmatrix) <- colnames(pathway)
223 
+colnames(PEAmatrix) <-  colnames(DataMatrix)
224 
+listIPA_pathways<-colnames(pathway)
225 
+for ( k in 1: nrow(PEAmatrix)){
226 
+  #k=1
227 
+  currentPathway <- colnames(pathway)[k]
228 
+  currentPathway_genes_list_common <- intersect(rownames(DataMatrix), currentPathway_genes<-pathway[,k])
229 
+  currentPathway_genes_list_commonMatrix <- DataMatrix[currentPathway_genes_list_common,]
230 
+  SumGenes <- colSums(currentPathway_genes_list_commonMatrix)
231 
+  AverageGenes <- SumGenes / length(currentPathway_genes_list_common)
232 
+  PEAmatrix[k,] <- AverageGenes
233 
+}
234 
+return(PEAmatrix)
235 
+}
236 
+
237 
+
238 
+
239 
+
240 
+
241 
+
242 
+
243 
+
244 
+
245 
+
246 
+
247 
+
248 
+#' @title For TCGA data get human pathway data and creates a measure of cross-talk among pathways
249 
+#' @description euc_dist_crtlk creates a matrix with euclidean distance for pairwise pathways
250 
+#' @param dataFilt TCGA matrix
251 
+#' @param pathway pathway data
252 
+#' @export
253 
+#' @return a matrix value for each pathway
254 
+#' @examples
255 
+#' score_euc_dista<-euc_dist_crtlk(dataFilt=tumo[,1:2],path)
256 
+euc_dist_crtlk <- function(dataFilt,pathway){
257 
+  PEAmatrix<-average(dataFilt,pathway)
258 
+  #step 5 distance
259 
+  # EUCLIDEA DISTANCE
260 
+  df=combn(rownames(PEAmatrix),2) # possibili relazioni tra i pathway
261 
+  df=t(df)
262 
+  ma_d<-matrix(0,nrow(df),ncol(PEAmatrix)) # creo matrix che conterr? le distanze
263 
+  colnames(ma_d)<-colnames(PEAmatrix) # colnames conterr? il nome dei pazienti
264 
+  for ( p in 1: ncol(PEAmatrix)){ # per ogni paziente
265 
+    patients <- (PEAmatrix)[,p]
266 
+    distance<-dist(patients) # calcolo distanza EUCLIDEA tra le possibile combinazioni
267 
+    ma_d[,p]<-distance
268 
+  }
269 
+  euc_dist<-cbind(df,ma_d) # inserisco label con le relazioni tra i pathway
270 
+  return(euc_dist)
271 
+}
272 
+
273 
+
274 
+
275 
+
276 
+#' @title For TCGA data get human pathway data and creates a measure of standard deviations among pathways
277 
+#' @description st_dv creates a matrix with standard deviation for pathways
278 
+#' @param DataMatrix TCGA matrix
279 
+#' @param pathway pathway data
280 
+#' @export
281 
+#' @return a matrix value for each pathway
282 
+#' @examples
283 
+#' stand_dev<-st_dv(DataMatrix=tumo[,1:2],pathway=path)
284 
+st_dv<-function(DataMatrix,pathway){
285 
+#DataMatrix<-dataFilt
286 
+
287 
+#dataFilt[ , "new.col"] <- gsub("\\|.*", "", rownames(dataFilt))
288 
+#DataMatrix<-dataFilt[which(dataFilt$new.col!="?"),]
289 
+#DataMatrix <- subset(DataMatrix, !duplicated(DataMatrix$new.col))
290 
+#rownames(DataMatrix)<-DataMatrix$new.col
291 
+#DataMatrix$new.col<-NULL
292 
+
293 
+PEAmatrix_sd <- matrix( 0, ncol(pathway),ncol(DataMatrix))
294 
+rownames(PEAmatrix_sd) <- colnames(pathway)
295 
+colnames(PEAmatrix_sd) <-  colnames(DataMatrix)
296 
+for ( k in 1: nrow(PEAmatrix_sd)){
297 
+  print(colnames(pathway)[k])
298 
+  currentPathway <- colnames(pathway)[k]
299 
+  currentPathway_genes_list_common <- intersect( rownames(DataMatrix), currentPathway_genes<-pathway[,k])
300 
+  currentPathway_genes_list_commonMatrix <- DataMatrix[currentPathway_genes_list_common,]
301 
+  stdev<-apply(currentPathway_genes_list_commonMatrix,2,sd) #deviazione standard dei pathway
302 
+  PEAmatrix_sd[k,] <- stdev
303 
+  }
304 
+return(PEAmatrix_sd)
305 
+}
306 
+
307 
+
308 
+
309 
+
310 
+
311 
+
312 
+#' @title For TCGA data get human pathway data and creates a measure of discriminating score among pathways
313 
+#' @description ds_score_crtlk creates a matrix with  discriminating score for pathways
314 
+#' @param dataFilt TCGA matrix
315 
+#' @param pathway pathway data
316 
+#' @export
317 
+#' @return a matrix value for each pathway
318 
+#' @examples
319 
+#' cross_talk_st_dv<-ds_score_crtlk(dataFilt=tumo[,1:2],pathway=path)
320 
+ds_score_crtlk<-function(dataFilt,pathway){
321 
+  PEAmatrix<-average(dataFilt,pathway)
322 
+  #step 5 distance
323 
+  # EUCLIDEA DISTANCE
324 
+  df=combn(rownames(PEAmatrix),2) # possibili relazioni tra i pathway
325 
+  df=t(df)
326 
+  ma_d<-matrix(0,nrow(df),ncol(PEAmatrix)) # creo matrix che conterr? le distanze
327 
+  colnames(ma_d)<-colnames(PEAmatrix) # colnames conterr? il nome dei pazienti
328 
+  for ( p in 1: ncol(PEAmatrix)){ # per ogni paziente
329 
+    patients <- (PEAmatrix)[,p]
330 
+    distance<-dist(patients) # calcolo distanza EUCLIDEA tra le possibile combinazioni
331 
+    ma_d[,p]<-distance
332 
+  }
333 
+  PEAmatrix_sd<-st_dv(dataFilt,pathway)
334 
+  df=combn(rownames(PEAmatrix_sd),2)
335 
+  df=t(df)
336 
+  ma<-matrix(0,nrow(df),ncol(PEAmatrix_sd)) # creo matrix che conterr? le somme delle dev st
337 
+  colnames(ma)<-colnames(PEAmatrix_sd) # colnames conterr? il nome dei pazienti
338 
+  for ( p in 1: ncol(PEAmatrix_sd)){ # per ogni paziente
339 
+    patients <- (PEAmatrix_sd)[,p]
340 
+    out <- apply(df, 1, function(x) sum(patients[x])) # calcolo somma delle dev standard tra le possibili combinazioni
341 
+    ma[,p]<-out
342 
+  }
343 
+  score<-ma_d/ma # discriminating score M1-M2/S1+S2
344 
+  score<- cbind(df,score)
345 
+return(score)
346 
+}
347 
+
348 
+
349 
+
350 
+#' @title SVM classification for each feature
351 
+#' @description svm class creates a list with auc value
352 
+#' @param TCGA_matrix gene expression matrix
353 
+#' @param nfs nfs split data into a training  and test set
354 
+#' @param tumour barcode samples for a class
355 
+#' @param normal barcode samples for another class
356 
+#' @export
357 
+#' @importFrom e1071 tune svm
358 
+#' @importFrom ROCR prediction performance
359 
+#' @importFrom  grDevices rainbow
360 
+#' @return a list with AUC value for pairwise pathway
361 
+#' @examples
362 
+#' nf <- 60
363 
+#' res_class<-svm_classification(TCGA_matrix=score_euc_dist,nfs=nf,
364 
+#' normal=colnames(norm[,1:10]),tumour=colnames(tumo[,1:10]))
365 
+svm_classification<-function(TCGA_matrix,tumour,normal,nfs){
366 
+  #library("e1071")
367 
+  #library(ROCR)
368 
+
369 
+  scoreMatrix <- as.data.frame(TCGA_matrix[,3:ncol(TCGA_matrix)])
370 
+  scoreMatrix <-as.data.frame(scoreMatrix)
371 
+  for( i in 1: ncol(scoreMatrix)){
372 
+    scoreMatrix[,i] <- as.numeric(as.character(scoreMatrix[,i]))
373 
+  }
374 
+
375 
+  TCGA_matrix[,1] <- gsub(" ", "_", TCGA_matrix[,1])
376 
+  d<-sub('_-_Homo_sapiens_*', '', TCGA_matrix[,1])
377 
+  #d_pr<-sub(')*', '', DataMatrix[,1])
378 
+
379 
+  d_pr<- gsub("(human)", "", d, fixed="TRUE")
380 
+  d_pr <- gsub("_", "", d_pr)
381 
+  d_pr <- gsub("-", "", d_pr)
382 
+
383 
+  TCGA_matrix[,2] <- gsub(" ", "_", TCGA_matrix[,2])
384 
+  d2<-sub('_-_Homo_sapiens_(human)*', '', TCGA_matrix[,2])
385 
+  d_pr2<- gsub("(human)", "", d2, fixed="TRUE")
386 
+  d_pr2 <- gsub("_", "", d_pr2)
387 
+  d_pr2 <- gsub("-", "", d_pr2)
388 
+
389 
+  PathwaysPair <- paste( as.matrix(d_pr), as.matrix(d_pr2),sep="_" )
390 
+
391 
+  rownames(scoreMatrix) <-PathwaysPair
392 
+
393 
+
394 
+  tDataMatrix<-as.data.frame(t(scoreMatrix))
395 
+  #tDataMatrix$Target[,1]<-0
396 
+
397 
+  tDataMatrix<-cbind(Target=0,tDataMatrix )
398 
+
399 
+  tum<-intersect(rownames(tDataMatrix),tumour)
400 
+  nor<-intersect(rownames(tDataMatrix),normal)
401 
+  #tDataMatrix$
402 
+
403 
+  Dataset_g1<-tDataMatrix[nor,]
404 
+  Dataset_g3<- tDataMatrix[tum,]
405 
+
406 
+
407 
+#training=read.table('C:/Users/UserInLab05/Desktop/trai.txt',header = TRUE)
408 
+#testset=read.table('C:/Users/UserInLab05/Desktop/test.txt',header = TRUE)
409 
+
410 
+  Dataset_g1$Target <- 0
411 
+  Dataset_g3$Target<-1
412 
+#Dataset_g3 <- Dataset_g3[Dataset_g3$Target <- 1, ]
413 
+
414 
+tab_g1_training <- sample(rownames(Dataset_g1),round(nrow(Dataset_g1) / 100 * nfs ))
415 
+tab_g3_training <- sample(rownames(Dataset_g3),round(nrow(Dataset_g3) / 100 * nfs ))
416 
+tab_g1_testing <- setdiff(rownames(Dataset_g1),tab_g1_training)
417 
+tab_g3_testing <- setdiff(rownames(Dataset_g3),tab_g3_training)
418 
+
419 
+FR<-intersect(rownames(Dataset_g1),tab_g1_training)
420 
+
421 
+#rownames(Dataset_g1)<-Dataset_g1[,1]
422 
+G1<-Dataset_g1[FR,]
423 
+
424 
+FR1<-intersect(rownames(Dataset_g3),tab_g3_training)
425 
+#rownames(Dataset_g3)<-Dataset_g3$ID
426 
+
427 
+G3<-Dataset_g3[FR1,]
428 
+training<-rbind(G1,G3)
429 
+
430 
+inter1<-intersect(rownames(Dataset_g1),tab_g1_testing)
431 
+#rownames(Dataset_g1)<-Dataset_g1$ID
432 
+
433 
+G1_testing<-Dataset_g1[inter1,]
434 
+
435 
+inter2<-intersect(rownames(Dataset_g3),tab_g3_testing)
436 
+#rownames(Dataset_g3)<-Dataset_g3$ID
437 
+G3_testing<-Dataset_g3[inter2,]
438 
+
439 
+testing<-rbind(G1_testing,G3_testing)
440 
+
441 
+x <- subset(training, select=-Target)
442 
+y <- training$Target
443 
+#testing[,2]<-NULL
444 
+z<-subset(testing, select=-Target)
445 
+
446 
+zi<-testing$Target
447 
+
448 
+auc.df<-list()
449 
+svm_model_after_tune_COMPL<-list()
450 
+for( k in 2: ncol(training)){
451 
+  print(colnames(training)[k])
452 
+
453 
+
454 
+
455 
+  svm_tune <- tune(svm, train.x=x, train.y=y,
456 
+                   kernel="radial", ranges=list(cost=10^(-1:2), gamma=c(.5,1,2)),cross=10)
457 
+  #print(svm_tune)
458 
+
459 
+  svm_model_after_tune <- svm(Target ~ ., data=training[,c(1,k)], kernel="radial", cost=svm_tune$best.parameters$cost, gamma=svm_tune$best.parameters$gamma,cross=10,probability = TRUE)
460 
+
461 
+
462 
+  #svm_model_after_tune <- svm(Target ~ ., data=training[,c(1,k)], kernel="radial", cost=svm_tune$best.parameters[1], gamma=svm_tune$best.parameters[2],cross=10,probability = TRUE)
463 
+  #summary(svm_model_after_tune)
464 
+
465 
+  j=k-1
466 
+  z2=z[,j]
467 
+  z3<-as.data.frame(z2)
468 
+  #rownames(z3)<-rownames(z)
469 
+  #colnames(z3)<-as.character(paste("X",j,sep = ""))
470 
+  colnames(z3)<-colnames(z)[j]
471 
+  #classifiersMatrix <- c(classifiersMatrix,svm_model_after_tune)
472 
+  pred <- predict(svm_model_after_tune,z3,decision.values=TRUE,cross=10)
473 
+
474 
+  #a<-table(pred,zi)
475 
+  svm.roc <- prediction(attributes(pred)$decision.values, zi)
476 
+  svm.auc <- performance(svm.roc, 'tpr', 'fpr')