1. StarBioTrek
  2. Commit 634693454fa3e27b91c7867040c61c9a62c24483
Browse code

fix2

git-svn-id: file:///home/git/hedgehog.fhcrc.org/bioconductor/branches/RELEASE_3_4/madman/Rpacks/StarBioTrek@125026 bc3139a8-67e5-0310-9ffc-ced21a209358

Claudia Cava authored on 12/12/2016 13:01:33
Showing 42 changed files
DESCRIPTION
History View file @ 6346934
1 1 
new file mode 100644
... ... 
@@ -0,0 +1,39 @@
1 
+Package: StarBioTrek
2 
+Type: Package
3 
+Title: StarBioTrek
4 
+Version: 1.0.1
5 
+Date: 12-12-2016
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 
+Description: This tool StarBioTrek presents some methodologies to measure pathway activity and cross-talk among pathways integrating also the information of network data.
20 
+License: GPL (>= 3)
21 
+biocViews: GeneRegulation,
22 
+    Network,
23 
+	Pathways,
24 
+	KEGG
25 
+Suggests:
26 
+    BiocStyle,
27 
+    knitr,
28 
+    rmarkdown,
29 
+    testthat,
30 
+	devtools,
31 
+	roxygen2,
32 
+	qgraph,
33 
+	png,
34 
+	grid
35 
+VignetteBuilder: knitr
36 
+LazyData: true
37 
+URL: https://github.com/claudiacava/StarBioTrek
38 
+BugReports: https://github.com/claudiacava/StarBioTrek/issues
39 
+RoxygenNote: 5.0.1
0 40 
\ No newline at end of file
NAMESPACE
History View file @ 6346934
1 41 
new file mode 100644
... ... 
@@ -0,0 +1,30 @@
1 
+# Generated by roxygen2: do not edit by hand
2 
+
3 
+export(GE_matrix)
4 
+export(SelectedSample)
5 
+export(average)
6 
+export(ds_score_crtlk)
7 
+export(euc_dist_crtlk)
8 
+export(getKEGGdata)
9 
+export(getNETdata)
10 
+export(list_path_net)
11 
+export(matrix_plot)
12 
+export(path_net)
13 
+export(plotting_cross_talk)
14 
+export(proc_path)
15 
+export(st_dv)
16 
+export(svm_classification)
17 
+importFrom(AnnotationDbi,as.list)
18 
+importFrom(AnnotationDbi,mappedkeys)
19 
+importFrom(KEGGREST,keggGet)
20 
+importFrom(KEGGREST,keggList)
21 
+importFrom(ROCR,performance)
22 
+importFrom(ROCR,prediction)
23 
+importFrom(SpidermiR,SpidermiRdownload_net)
24 
+importFrom(SpidermiR,SpidermiRprepare_NET)
25 
+importFrom(SpidermiR,SpidermiRquery_spec_networks)
26 
+importFrom(SpidermiR,SpidermiRquery_species)
27 
+importFrom(e1071,svm)
28 
+importFrom(e1071,tune)
29 
+importFrom(grDevices,rainbow)
30 
+importFrom(org.Hs.eg.db,org.Hs.egSYMBOL2EG)
NEWS
History View file @ 6346934
0 31 
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 @ 6346934
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 @ 6346934
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_gly(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 @ 6346934
0 208 
new file mode 100644
... ... 
@@ -0,0 +1,539 @@
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 
+ac<-paste("N-Glycan biosynthesis", species)
101 
+ad<-paste("Various types of N-glycan biosynthesis", species)
102 
+ae<-paste("Mucin type O-Glycan biosynthesis", species)
103 
+af<-paste("Other types of O-glycan biosynthesis", species)
104 
+ag<-paste("Glycosaminoglycan biosynthesis - CS/DS", species)
105 
+ah<-paste("Glycosaminoglycan biosynthesis - HS/Hep", species)
106 
+ai<-paste("Glycosaminoglycan biosynthesis - KS", species)
107 
+al<-paste("Glycosaminoglycan degradation", species)
108 
+am<-paste("Glycosylphosphatidylinositol(GPI)-anchor biosynthesis", species)
109 
+an<-paste("Glycosphingolipid biosynthesis - lacto and neolacto series", species)
110 
+ao<-paste("Glycosphingolipid biosynthesis - globo series", species)
111 
+ap<-paste("Glycosphingolipid biosynthesis - ganglio series", species)
112 
+aq<-paste("Lipopolysaccharide biosynthesis", species)
113 
+ar<-paste("Peptidoglycan biosynthesis", species)
114 
+as<-paste("Other glycan degradation", species)
115 
+mer<-c(ac,ad,ae,af,ag,ah,ai,al,am,an,ao,ap,aq,ar,as)
116 
+return(mer)
117 
+}
118 
+
119 
+
120 
+
121 
+select_path_cofa<-function(Cofa_vita_met){
122 
+  species<-c("- Homo sapiens (human)")
123 
+ac<-paste("Thiamine metabolism", species)
124 
+ad<-paste("Riboflavin metabolism", species)
125 
+ae<-paste("Vitamin B6 metabolism", species)
126 
+af<-paste("Nicotinate and nicotinamide metabolism", species)
127 
+ag<-paste("Pantothenate and CoA biosynthesis", species)
128 
+ah<-paste("Biotin metabolism", species)
129 
+ai<-paste("Lipoic acid metabolism", species)
130 
+al<-paste("Folate biosynthesis", species)
131 
+am<-paste("One carbon pool by folate", species)
132 
+an<-paste("Retinol metabolism", species)
133 
+ao<-paste("Porphyrin and chlorophyll metabolism", species)
134 
+ap<-paste("Ubiquinone and other terpenoid-quinone biosynthesis", species)
135 
+mer<-c(ac,ad,ae,af,ag,ah,ai,al,am,an,ao,ap)
136 
+return(mer)
137 
+}
138 
+
139 
+select_path_transc<-function(Transcription){
140 
+  species<-c("- Homo sapiens (human)")
141 
+ac<-paste("RNA polymerase", species)
142 
+ad<-paste("Basal transcription factors", species)
143 
+ae<-paste("Spliceosome", species)
144 
+af<-paste("Transcription factors", species)
145 
+ag<-paste("Transcription machinery", species)
146 
+mer<-c(ac,ad,ae,af,ag)
147 
+return(mer)
148 
+}
149 
+
150 
+
151 
+
152 
+select_path_transl<-function(Translation){
153 
+  species<-c("- Homo sapiens (human)")
154 
+ac<-paste("Ribosome", species)
155 
+ad<-paste("Aminoacyl-tRNA biosynthesis", species)
156 
+ae<-paste("RNA transport", species)
157 
+af<-paste("mRNA surveillance pathway", species)
158 
+ag<-paste("Ribosome biogenesis in eukaryotes", species)
159 
+ah<-paste("Ribosomal proteins", species)
160 
+ai<-paste("Ribosome biogenesis", species)
161 
+al<-paste("Transfer RNA biogenesis", species)
162 
+am<-paste("Translation factors", species)
163 
+mer<-c(ac,ad,ae,af,ag,ah,ai,al,am)
164 
+return(mer)
165 
+}
166 
+
167 
+select_path_fold<-function(Folding_sorting_and_degradation){
168 
+  species<-c("- Homo sapiens (human)")
169 
+ac<-paste("Protein export", species)
170 
+ad<-paste("Protein processing in endoplasmic reticulum", species)
171 
+ae<-paste("SNARE interactions in vesicular transport", species)
172 
+af<-paste("Ubiquitin mediated proteolysis", species)
173 
+ag<-paste("Sulfur relay system", species)
174 
+ah<-paste("RNA degradation", species)
175 
+ai<-paste("Chaperones and folding catalysts", species)
176 
+al<-paste("SNAREs", species)
177 
+am<-paste("Ubiquitin system", species)
178 
+an<-paste("Proteasome", species)
179 
+mer<-c(ac,ad,ae,af,ag,ah,ai,al,am,an)
180 
+return(mer)
181 
+}
182 
+
183 
+
184 
+
185 
+
186 
+select_path_repl<-function(Replication_and_repair){
187 
+  species<-c("- Homo sapiens (human)")
188 
+ac<-paste("DNA replication", species)
189 
+ad<-paste("Base excision repair", species)
190 
+ae<-paste("Nucleotide excision repair", species)
191 
+af<-paste("Mismatch repair", species)
192 
+ag<-paste("Homologous recombination", species)
193 
+ah<-paste("Non-homologous end-joining", species)
194 
+ai<-paste("Fanconi anemia pathway", species)
195 
+al<-paste("DNA replication proteins", species)
196 
+am<-paste("Chromosome", species)
197 
+an<-paste("DNA repair and recombination", species)
198 
+ao<-paste("proteins", species)
199 
+mer<-c(ac,ad,ae,af,ag,ah,ai,al,am,an,ao)
200 
+return(mer)
201 
+}
202 
+
203 
+
204 
+
205 
+select_path_sign<-function(Signal_transduction){
206 
+  species<-c("- Homo sapiens (human)")
207 
+a<-paste("Ras signaling pathway", species)
208 
+b<-paste("Rap1 signaling pathway", species)
209 
+c<-paste("MAPK signaling pathway", species)
210 
+d<-paste("ErbB signaling pathway", species)
211 
+e<-paste("Wnt signaling pathway", species)
212 
+f<-paste("Notch signaling pathway", species)
213 
+g<-paste("Hedgehog signaling pathway", species)
214 
+h<-paste("TGF-beta signaling pathway", species)
215 
+i<-paste("Hippo signaling pathway", species)
216 
+l<-paste("VEGF signaling pathway", species)
217 
+m<-paste("Jak-STAT signaling pathway", species)
218 
+n<-paste("NF-kappa B signaling pathway", species)
219 
+o<-paste("TNF signaling pathway", species)
220 
+p<-paste("HIF-1 signaling pathway", species)
221 
+q<-paste("FoxO signaling pathway", species)
222 
+r<-paste("Calcium signaling pathway", species)
223 
+s<-paste("Phosphatidylinositol signaling system", species)
224 
+t<-paste("Phospholipase D signaling pathway", species)
225 
+v<-paste("Sphingolipid signaling pathway", species)
226 
+u<-paste("cAMP signaling pathway", species)
227 
+z<-paste("cGMP-PKG signaling pathway", species)
228 
+ab<-paste("PI3K-Akt signaling pathway", species)
229 
+ac<-paste("AMPK signaling pathway", species)
230 
+ad<-paste("mTOR signaling pathway", species)
231 
+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)
232 
+return(mer)
233 
+}
234 
+
235 
+
236 
+select_path_sign_mol<-function(Signaling_molecules_and_interaction){
237 
+  species<-c("- Homo sapiens (human)")
238 
+a<-paste("Neuroactive ligand-receptor interaction", species)
239 
+b<-paste("Cytokine-cytokine receptor interaction", species)
240 
+c<-paste("ECM-receptor interaction", species)
241 
+d<-paste("Cell adhesion molecules (CAMs)", species)
242 
+mer<-c(a,b,c,d)
243 
+return(mer)
244 
+}
245 
+
246 
+
247 
+select_path_transp_ca<-function(Transport_and_catabolism){
248 
+  species<-c("- Homo sapiens (human)")
249 
+a<-paste("Endocytosis", species)
250 
+b<-paste("Phagosome", species)
251 
+c<-paste("Lysosome", species)
252 
+d<-paste("Peroxisome", species)
253 
+e<-paste("Regulation of autophagy", species)
254 
+mer<-c(a,b,c,d,e)
255 
+return(mer)
256 
+}
257 
+
258 
+select_path_cell_grow<-function(Cell_growth_and_death){
259 
+  species<-c("- Homo sapiens (human)")
260 
+  a<-paste("Cell cycle", species)
261 
+b<-paste("Apoptosis", species)
262 
+c<-paste("p53 signaling pathway", species)
263 
+mer<-c(a,b,c)
264 
+return(mer)
265 
+}
266 
+
267 
+
268 
+select_path_cell_comm<-function(Cellular_community){
269 
+  species<-c("- Homo sapiens (human)")
270 
+  a<-paste("Focal adhesion", species)
271 
+b<-paste("Adherens junction", species)
272 
+c<-paste("Tight junction", species)
273 
+d<-paste("Gap junction", species)
274 
+e<-paste("Signaling pathways regulating pluripotency of stem cells ", species)
275 
+mer<-c(a,b,c,d,e)
276 
+return(mer)
277 
+}
278 
+
279 
+
280 
+select_path_imm_syst<-function(Immune_system){
281 
+  species<-c("- Homo sapiens (human)")
282 
+a<-paste("Hematopoietic cell lineage", species)
283 
+b<-paste("Complement and coagulation cascades", species)
284 
+c<-paste("Platelet activation", species)
285 
+d<-paste("Toll-like receptor signaling pathway", species)
286 
+e<-paste("Toll and Imd signaling pathway", species)
287 
+f<-paste("NOD-like receptor signaling pathway", species)
288 
+g<-paste("RIG-I-like receptor signaling pathway", species)
289 
+h<-paste("Cytosolic DNA-sensing pathway", species)
290 
+i<-paste("Natural killer cell mediated cytotoxicity", species)
291 
+l<-paste("Antigen processing and presentation", species)
292 
+m<-paste("T cell receptor signaling pathway", species)
293 
+n<-paste("B cell receptor signaling pathway", species)
294 
+o<-paste("Fc epsilon RI signaling pathway", species)
295 
+p<-paste("Fc gamma R-mediated phagocytosis", species)
296 
+q<-paste("Leukocyte transendothelial migration", species)
297 
+r<-paste("Intestinal immune network for IgA production", species)
298 
+s<-paste("Chemokine signaling pathway", species)
299 
+
300 
+mer<-c(a,b,c,d,e,f,g,h,i,l,m,n,o,p,q,r,s)
301 
+return(mer)
302 
+}
303 
+
304 
+
305 
+
306 
+
307 
+select_path_end_syst<-function(Endocrine_system){
308 
+  species<-c("- Homo sapiens (human)")
309 
+a<-paste("Insulin secretion", species)
310 
+b<-paste("Insulin signaling pathway", species)
311 
+c<-paste("Glucagon signaling pathway", species)
312 
+d<-paste("Regulation of lipolysis in adipocytes", species)
313 
+e<-paste("Adipocytokine signaling pathway", species)
314 
+f<-paste("PPAR signaling pathway", species)
315 
+g<-paste("GnRH signaling pathway", species)
316 
+h<-paste("Ovarian steroidogenesis", species)
317 
+i<-paste("Estrogen signaling pathway", species)
318 
+l<-paste("Progesterone-mediated oocyte maturation", species)
319 
+m<-paste("Prolactin signaling pathway", species)
320 
+n<-paste("Oxytocin signaling pathway", species)
321 
+o<-paste("Thyroid hormone synthesis", species)
322 
+p<-paste("Thyroid hormone signaling pathway", species)
323 
+q<-paste("Melanogenesis", species)
324 
+r<-paste("Renin secretion", species)
325 
+s<-paste("Renin-angiotensin system", species)
326 
+t<-paste("Aldosterone synthesis and secretion", species)
327 
+
328 
+
329 
+mer<-c(a,b,c,d,e,f,g,h,i,l,m,n,o,p,q,r,s,t)
330 
+return(mer)
331 
+}
332 
+
333 
+
334 
+select_path_circ_syst<-function(Circulatory_system){
335 
+  species<-c("- Homo sapiens (human)")
336 
+  a<-paste("Cardiac muscle contraction", species)
337 
+b<-paste("Adrenergic signaling in cardiomyocytes", species)
338 
+c<-paste("Vascular smooth muscle contraction", species)
339 
+mer<-c(a,b,c)
340 
+return(mer)
341 
+}
342 
+
343 
+
344 
+select_path_dig_syst<-function(Digestive_system){
345 
+  species<-c("- Homo sapiens (human)")
346 
+  a<-paste("Salivary secretion", species)
347 
+b<-paste("Gastric acid secretion", species)
348 
+c<-paste("Pancreatic secretion", species)
349 
+d<-paste("Bile secretion", species)
350 
+e<-paste("Carbohydrate digestion and absorption", species)
351 
+f<-paste("Protein digestion and absorption", species)
352 
+g<-paste("Fat digestion and absorption", species)
353 
+h<-paste("Vitamin digestion and absorption", species)
354 
+i<-paste("Mineral absorption", species)
355 
+
356 
+mer<-c(a,b,c,d,e,f,g,h,i)
357 
+return(mer)
358 
+}
359 
+
360 
+
361 
+
362 
+select_path_exc_syst<-function(Excretory_system){
363 
+  species<-c("- Homo sapiens (human)")
364 
+  a<-paste("Vasopressin-regulated water reabsorption", species)
365 
+b<-paste("Aldosterone-regulated sodium reabsorption", species)
366 
+c<-paste("Endocrine and other factor-regulated calcium reabsorption", species)
367 
+d<-paste("Proximal tubule bicarbonate reclamation", species)
368 
+e<-paste("Collecting duct acid secretion", species)
369 
+
370 
+
371 
+mer<-c(a,b,c,d,e)
372 
+return(mer)
373 
+}
374 
+
375 
+
376 
+select_path_ner_syst<-function(Nervous_system){
377 
+  species<-c("- Homo sapiens (human)")
378 
+a<-paste("Glutamatergic synapse", species)
379 
+b<-paste("GABAergic synapse", species)
380 
+c<-paste("Cholinergic synapse", species)
381 
+d<-paste("Dopaminergic synapse", species)
382 
+e<-paste("Serotonergic synapse", species)
383 
+f<-paste("Long-term potentiation", species)
384 
+g<-paste("Long-term depression", species)
385 
+h<-paste("Retrograde endocannabinoid signaling", species)
386 
+i<-paste("Synaptic vesicle cycle", species)
387 
+l<-paste("Neurotrophin signaling pathway", species)
388 
+
389 
+mer<-c(a,b,c,d,e,f,g,h,i,l)
390 
+return(mer)
391 
+}
392 
+
393 
+
394 
+select_path_sens_syst<-function(Sensory_system){
395 
+  species<-c("- Homo sapiens (human)")
396 
+  a<-paste("Phototransduction", species)
397 
+b<-paste("Olfactory transduction", species)
398 
+c<-paste("Taste transduction", species)
399 
+d<-paste("Inflammatory mediator regulation of TRP channels", species)
400 
+mer<-c(a,b,c,d)
401 
+return(mer)
402 
+}
403 
+
404 
+
405 
+
406 
+#' @title Select the class of TCGA data
407 
+#' @description select two labels from ID barcode
408 
+#' @param Dataset gene expression matrix
409 
+#' @param typesample the labels of the samples (e.g. tumor,normal)
410 
+#' @export
411 
+#' @return a gene expression matrix of the samples with specified label
412 
+#' @examples
413 
+#' tumo<-SelectedSample(Dataset=Data_CANCER_normUQ_filt,typesample="tumor")[,2]
414 
+SelectedSample <- function(Dataset,typesample){
415 
+  if( typesample =="tumor"){
416 
+    Dataset <- Dataset[,which( as.numeric(substr(colnames(Dataset), 14, 15)) == 01) ]
417 
+  }
418 
+
419 
+  if( typesample =="normal"){
420 
+    Dataset <- Dataset[,which( as.numeric(substr(colnames(Dataset), 14, 15)) >= 10) ]
421 
+  }
422 
+
423 
+  return(Dataset)
424 
+
425 
+}
426 
+
427 
+
428 
+#' @title Select the class of TCGA data
429 
+#' @description select two labels from ID barcode
430 
+#' @param cutoff cut-off for AUC value
431 
+#' @param auc.df list of AUC value
432 
+#' @return a gene expression matrix with only pairwise pathway with a particular cut-off
433 
+select_class<-function(auc.df,cutoff){
434 
+ds<-do.call("rbind", auc.df)
435 
+tmp_ordered <- as.data.frame(ds[order(ds,decreasing=TRUE),])
436 
+colnames(tmp_ordered)<-'pathway'
437 
+er<-as.data.frame(tmp_ordered$pathway>cutoff)
438 
+ase<-tmp_ordered[tmp_ordered$pathway>cutoff,]
439 
+rownames(er)<-rownames(tmp_ordered)
440 
+er[,2]<-tmp_ordered$pathway
441 
+lipid_metabolism<-er[1:length(ase),]
442 
+return(lipid_metabolism)
443 
+}
444 
+
445 
+
446 
+
447 
+
448 
+#' @title Process matrix TCGA data after the selection of pairwise pathway
449 
+#' @description processing gene expression matrix
450 
+#' @param measure matrix with measure of cross-talk among pathways
451 
+#' @param list_perf output of the function select_class
452 
+#' @return a gene expression matrix for case study 1
453 
+process_matrix<-function(measure,list_perf){
454 
+scoreMatrix <- as.data.frame(measure[,3:ncol(measure)])
455 
+for( i in 1: ncol(scoreMatrix)){
456 
+  scoreMatrix[,i] <- as.numeric(as.character(scoreMatrix[,i]))
457 
+}
458 
+measure[,1] <- gsub(" ", "_", measure[,1])
459 
+d<-sub('_-_Homo_sapiens_*', '', measure[,1])
460 
+d_pr<- gsub("(human)", "", d, fixed="TRUE")
461 
+d_pr <- gsub("_", "", d_pr)
462 
+d_pr <- gsub("-", "", d_pr)
463 
+measure[,2] <- gsub(" ", "_", measure[,2])
464 
+d2<-sub('_-_Homo_sapiens_(human)*', '', measure[,2])
465 
+d_pr2<- gsub("(human)", "", d2, fixed="TRUE")
466 
+d_pr2 <- gsub("_", "", d_pr2)
467 
+d_pr2 <- gsub("-", "", d_pr2)
468 
+PathwaysPair <- paste( as.matrix(d_pr), as.matrix(d_pr2),sep="_" )
469 
+rownames(scoreMatrix) <-PathwaysPair
470 
+intera<-intersect(rownames(scoreMatrix),rownames(list_perf))
471 
+path_bestlipd<-scoreMatrix[intera,]
472 
+return(path_bestlipd)
473 
+}
474 
+
475 
+
476 
+
477 
+process_matrix_cell_process<-function(measure_cell_process){
478 
+score__cell_grow_d <- as.data.frame(measure_cell_process[,3:ncol(measure_cell_process)])
479 
+for( i in 1: ncol(score__cell_grow_d)){
480 
+  score__cell_grow_d[,i] <- as.numeric(as.character(score__cell_grow_d[,i]))
481 
+}
482 
+
483 
+measure_cell_process[,1] <- gsub(" ", "_", measure_cell_process[,1])
484 
+d<-sub('_-_Homo_sapiens_*', '', measure_cell_process[,1])
485 
+
486 
+d_pr<- gsub("(human)", "", d, fixed="TRUE")
487 
+d_pr <- gsub("_", "", d_pr)
488 
+d_pr <- gsub("-", "", d_pr)
489 
+
490 
+measure_cell_process[,2] <- gsub(" ", "_", measure_cell_process[,2])
491 
+d2<-sub('_-_Homo_sapiens_(human)*', '', measure_cell_process[,2])
492 
+d_pr2<- gsub("(human)", "", d2, fixed="TRUE")
493 
+d_pr2 <- gsub("_", "", d_pr2)
494 
+d_pr2 <- gsub("-", "", d_pr2)
495 
+
496 
+PathwaysPair <- paste( as.matrix(d_pr), as.matrix(d_pr2),sep="_" )
497 
+rownames(score__cell_grow_d) <-PathwaysPair
498 
+return(score__cell_grow_d)
499 
+}
500 
+
501 
+
502 
+#' @title Get human KEGG pathway data.
503 
+#' @description getKEGGdata creates a data frame with human KEGG pathway. Columns are the pathways and rows the genes inside those pathway
504 
+#' @param mer  output for example of select_path_carb
505 
+#' @export
506 
+#' @importFrom KEGGREST keggList
507 
+#' @return dataframe with human pathway data
508 
+proc_path<-function(mer){
509 
+pathways.list <- keggList("pathway", "hsa")## returns the list of human pathways
510 
+common<-intersect(pathways.list,mer)
511 
+lo<-list()
512 
+for (i in 1:length(pathways.list)){
513 
+  if (length(intersect(pathways.list[[i]],common)!=0)){
514 
+    lo[[i]]<-pathways.list[[i]]
515 
+    names(lo)[[i]]<-names(pathways.list)[[i]]
516 
+  }
517 
+}
518 
+pathways.list<-lo[lapply(lo,length)!=0]
519 
+pathway.codes <- sub("path:", "", names(pathways.list))
520 
+b<-do.call("rbind", pathways.list)
521 
+list_pathkegg<-list(pathway.codes,b)
522 
+return(list_pathkegg)
523 
+}
524 
+
525 
+
526 
+overlap <- function(net_type,x,currentPathway_genes){
527 
+  de<-net_type[which(net_type$m_shar_pro==x),]
528 
+  fr<-intersect(de$m2_shar_pro,currentPathway_genes)
529 
+  go=list()
530 
+  if(length(fr)!=0)    {
531 
+    for (i in 1:length(fr)){
532 
+      de2<-de[which(de$m2_shar_pro==fr[i]),]
533 
+      go[[i]]<-de2
534 
+    }
535 
+  }
536 
+  dst<-do.call("rbind", go)
537 
+  return(dst)
538 
+}
539 
+
R/path_star.R
History View file @ 6346934
0 540 
new file mode 100644
... ... 
@@ -0,0 +1,515 @@
1 
+#' @title Get human KEGG pathway data and network data in order to define the common gene.
2 
+#' @description path_net creates a list of network data for each human pathway. The network data will be generated when interacting genes belong to that pathway.
3 
+#' @param net_type  network data as provided by getNETdata
4 
+#' @param pathway  pathway data as provided by getKEGGdata
5 
+#' @export
6 
+#' @return a list of network data for each pathway (interacting genes belong to that pathway)
7 
+#' @examples
8 
+#' lista_net<-path_net(pathway=path,net_type=netw)
9 
+path_net<-function(pathway,net_type){
10 
+  lista_int<-list()
11 
+  for (k in 1:ncol(pathway)){
12 
+    #k=1
13 
+    print(paste(k,"PATHWAY",colnames(pathway)[k]))
14 
+    currentPathway_genes<-pathway[,k]
15 
+    common1 <- intersect( net_type$m_shar_pro, currentPathway_genes)
16 
+    common2 <- intersect( net_type$m2_shar_pro, currentPathway_genes)
17 
+    if (length(common1)==0 & length(common2)==0 ){
18 
+      mago2<-character(length = 0)
19 
+    }
20 
+    if (length(common1)!=0 | length(common2)!=0 ){
21 
+      b=list()
22 
+      for (i in 1:length(common1)){
23 
+        x<-common1[i]
24 
+        n<-overlap(net_type,x,currentPathway_genes)
25 
+        b[[i]]<-n
26 
+      }
27 
+      v<-do.call("rbind", b)
28 
+      c=list()
29 
+      for (i in 1:length(common2)){
30 
+        x<-common1[i]
31 
+        n<-overlap(net_type,x,currentPathway_genes)
32 
+        c[[i]]<-n
33 
+      }
34 
+      v2<-do.call("rbind", b)
35 
+      mago<-rbind(v,v2)
36 
+      mago2<-mago[!duplicated(mago), ]
37 
+    }
38 
+
39 
+    if (length(mago2)!=0){
40 
+      lista_int[[k]]<-mago2
41 
+    }
42 
+    if (length(mago2)==0){
43 
+      lista_int[[k]]<-"0"}
44 
+
45 
+    names(lista_int)[k]<-colnames(pathway)[k]
46 
+  }
47 
+  return(lista_int)
48 
+}
49 
+
50 
+
51 
+#' @title Get human KEGG pathway data and output of path_net in order to define the common genes.
52 
+#' @description list_path_net creates a list of interacting genes for each human pathway.
53 
+#' @param lista_net  output of path_net
54 
+#' @param pathway  pathway data as provided by getKEGGdata
55 
+#' @export
56 
+#' @return a list of genes for each pathway (interacting genes belong to that pathway)
57 
+#' @examples
58 
+#' lista_netw<-path_net(pathway=path,net_type=netw)
59 
+#' list_path<-list_path_net(lista_net=lista_netw,pathway=path)
60 
+list_path_net<-function(lista_net,pathway){
61 
+v=list()
62 
+bn=list()
63 
+for (j in 1:length(lista_net)){
64 
+  cf<-lista_net[[j]]
65 
+  i <- sapply(cf, is.factor)
66 
+  cf[i] <- lapply(cf[i], as.character)
67 
+  m<-c(cf$m_shar_pro)
68 
+  m2<-c(cf$m2_shar_pro)
69 
+  s<-c(m,m2)
70 
+  fr<- unique(s)
71 
+  n<-as.data.frame(fr)
72 
+  if(length(n)==0){
73 
+    v[[j]]<-NULL
74 
+
75 
+  }
76 
+  if(length(n)!=0){
77 
+  i <- sapply(n, is.factor)
78 
+  n[i] <- lapply(n[i], as.character)
79 
+  #for (k in  1:ncol(pathway)){
80 
+  if (length(intersect(n$fr,pathway[,j]))==nrow(n)){
81 
+    print(paste("List of genes interacting in the same pathway:",colnames(pathway)[j]))
82 
+    aa<-intersect(n$fr,pathway[,j])
83 
+    v[[j]]<-aa
84 
+    names(v)[j]<-colnames(pathway)[j]
85 
+  }
86 
+}}
87 
+return(v)}
88 
+
89 
+
90 
+
91 
+
92 
+#' @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 .
93 
+#' @description GE_matrix creates a matrix of gene expression for pathways given by the user.
94 
+#' @param DataMatrix  gene expression matrix (eg.TCGA data)
95 
+#' @param pathway  pathway data as provided by getKEGGdata
96 
+#' @export
97 
+#' @return a matrix for each pathway ( gene expression level belong to that pathway)
98 
+#' @examples
99 
+#' list_path_gene<-GE_matrix(DataMatrix=tumo[,1:2],pathway=path)
100 
+GE_matrix<-function(DataMatrix,pathway) {
101 
+  path_name<-sub(' ', '_',colnames(pathway))
102 
+d_pr<- gsub(" - Homo sapiens (human)", "", path_name, fixed="TRUE")
103 
+colnames(pathway)<-d_pr
104 
+#zz<-as.data.frame(rowMeans(DataMatrix))
105 
+zz<-as.data.frame(DataMatrix)
106 
+v<-list()
107 
+for ( k in 1: ncol(pathway)){
108 
+  #k=2
109 
+  if (length(intersect(rownames(zz),pathway[,k])!=0)){
110 
+    print(colnames(path)[k])
111 
+  currentPathway_genes_list_common <- intersect(rownames(zz), currentPathway_genes<-pathway[,k])
112 
+  currentPathway_genes_list_commonMatrix <- as.data.frame(zz[currentPathway_genes_list_common,])
113 
+  rownames(currentPathway_genes_list_commonMatrix)<-currentPathway_genes_list_common
114 
+  v[[k]]<- currentPathway_genes_list_commonMatrix
115 
+  names(v)[k]<-colnames(pathway)[k]
116 
+  }
117 
+}
118 
+#PEAmatrix <- matrix( 0,nrow(DataMatrix),ncol(pathway))
119 
+#rownames(PEAmatrix) <- as.factor(rownames(DataMatrix))
120 
+#colnames(PEAmatrix) <-  as.factor(colnames(pathway))
121 
+#for (i in 1:length(v)){
122 
+#PEAmatrix[v[[i]],i]<-zz[v[[i]],]
123 
+#}
124 
+#PEAmatrix<-PEAmatrix[which(rowSums(PEAmatrix) > 0),]
125 
+return(v)
126 
+}
127 
+
128 
+
129 
+
130 
+#' @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 .
131 
+#' @description GE_matrix creates a matrix of mean gene expression for pathways given by the user.
132 
+#' @param DataMatrix  gene expression matrix (eg.TCGA data)
133 
+#' @param pathway  pathway data as provided by getKEGGdata
134 
+#' @export
135 
+#' @return a matrix for each pathway (mean gene expression level belong to that pathway)
136 
+#' @examples
137 
+#' list_path_plot<-matrix_plot(DataMatrix=tumo[,1:2],pathway=path)
138 
+matrix_plot<-function(DataMatrix,pathway) {
139 
+  path_name<-sub(' ', '_',colnames(pathway))
140 
+  d_pr<- gsub(" - Homo sapiens (human)", "", path_name, fixed="TRUE")
141 
+  colnames(pathway)<-d_pr
142 
+  zz<-as.data.frame(rowMeans(DataMatrix))
143 
+  v<-list()
144 
+  for ( k in 1: ncol(pathway)){
145 
+    #k=2
146 
+    if (length(intersect(rownames(zz),pathway[,k])!=0)){
147 
+      print(colnames(path)[k])
148 
+      currentPathway_genes_list_common <- intersect(rownames(zz), currentPathway_genes<-pathway[,k])
149 
+      currentPathway_genes_list_commonMatrix <- as.data.frame(zz[currentPathway_genes_list_common,])
150 
+      rownames(currentPathway_genes_list_commonMatrix)<-currentPathway_genes_list_common
151 
+      v[[k]]<- currentPathway_genes_list_common
152 
+      names(v)[k]<-colnames(pathway)[k]
153 
+    }
154 
+  }
155 
+  PEAmatrix <- matrix( 0,nrow(DataMatrix),ncol(pathway))
156 
+  rownames(PEAmatrix) <- as.factor(rownames(DataMatrix))
157 
+  colnames(PEAmatrix) <-  as.factor(colnames(pathway))
158 
+  for (i in 1:length(v)){
159 
+  PEAmatrix[v[[i]],i]<-zz[v[[i]],]
160 
+  }
161 
+  PEAmatrix<-PEAmatrix[which(rowSums(PEAmatrix) > 0),]
162 
+  return(PEAmatrix)
163 
+}
164 
+
165 
+
166 
+
167 
+
168 
+
169 
+
170 
+
171 
+
172 
+
173 
+
174 
+
175 
+
176 
+
177 
+#' @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 .
178 
+#' @description plotting_matrix creates a matrix of gene expression for pathways given by the user.
179 
+#' @param DataMatrix  gene expression matrix (eg.TCGA data)
180 
+#' @param pathway  pathway data as provided by getKEGGdata
181 
+#' @param path_matrix  output of the function matrix_plot
182 
+#' @export
183 
+#' @return a plot for pathway cross talk
184 
+#' @examples
185 
+#' mt<-plotting_cross_talk(DataMatrix=tumo[,1:2],pathway=path,path_matrix=list_path_plot)
186 
+plotting_cross_talk<-function(DataMatrix,pathway,path_matrix){
187 
+  zz<-as.data.frame(rowMeans(DataMatrix))
188 
+  v<-list()
189 
+  for ( k in 1: ncol(pathway)){
190 
+    path_name<-sub(' ', '_',colnames(pathway))
191 
+    d_pr<- gsub(" - Homo sapiens (human)", "", path_name, fixed="TRUE")
192 
+    colnames(pathway)<-d_pr
193 
+    if (length(intersect(rownames(zz),pathway[,k])!=0)){
194 
+      print(colnames(path)[k])
195 
+      currentPathway_genes_list_common <- intersect(rownames(zz), currentPathway_genes<-pathway[,k])
196 
+      currentPathway_genes_list_commonMatrix <- as.data.frame(zz[currentPathway_genes_list_common,])
197 
+      rownames(currentPathway_genes_list_commonMatrix)<-currentPathway_genes_list_common
198 
+      v[[k]]<- as.factor(currentPathway_genes_list_common)
199 
+      names(v)[k]<-colnames(pathway)[k]
200 
+    }
201 
+  }
202 
+  vv<-list()
203 
+  mi<-t(path_matrix)
204 
+
205 
+  dc<-cor(mi)
206 
+  for ( k in 1: length(v)){
207 
+    currentPathway_genes_list_common <- intersect(rownames(dc), v[[k]])
208 
+    a<-match(currentPathway_genes_list_common,rownames(dc))
209 
+    vv[[k]]<- a
210 
+    names(vv)[k]<-colnames(pathway)[k]
211 
+  }
212 
+  list_plt=list(corr=dc,gruppi=vv)
213 
+ #r<-qgraph(list_plt$corr, groups=list_plt$gruppi, mar=c(1,1,1,1),minimum=0.6)
214 
+  return(list_plt)
215 
+}
216 
+
217 
+
218 
+
219 
+
220 
+#' @title For TCGA data get human pathway data and creates a matrix with the average of genes for each pathway.
221 
+#' @description average creates a matrix with a summarized value for each pathway
222 
+#' @param dataFilt TCGA matrix
223 
+#' @param pathway pathway data
224 
+#' @export
225 
+#' @return a matrix value for each pathway
226 
+#' @examples
227 
+#' score_mean<-average(dataFilt=tumo[,1:2],path)
228 
+average<-function(dataFilt,pathway){
229 
+  DataMatrix<-dataFilt
230 
+  #dataFilt[ , "new.col"] <- gsub("\\|.*", "", rownames(dataFilt))
231 
+  #DataMatrix<-dataFilt[which(dataFilt$new.col!="?"),]
232 
+  #DataMatrix <- subset(DataMatrix, !duplicated(DataMatrix$new.col))
233 
+  #rownames(DataMatrix)<-DataMatrix$new.col
234 
+  #DataMatrix$new.col<-NULL
235 
+
236 
+PEAmatrix <- matrix( 0, ncol(pathway),ncol(DataMatrix))
237 
+rownames(PEAmatrix) <- colnames(pathway)
238 
+colnames(PEAmatrix) <-  colnames(DataMatrix)
239 
+listIPA_pathways<-colnames(pathway)
240 
+for ( k in 1: nrow(PEAmatrix)){
241 
+  #k=1
242 
+  currentPathway <- colnames(pathway)[k]
243 
+  currentPathway_genes_list_common <- intersect(rownames(DataMatrix), currentPathway_genes<-pathway[,k])
244 
+  currentPathway_genes_list_commonMatrix <- DataMatrix[currentPathway_genes_list_common,]
245 
+  SumGenes <- colSums(currentPathway_genes_list_commonMatrix)
246 
+  AverageGenes <- SumGenes / length(currentPathway_genes_list_common)
247 
+  PEAmatrix[k,] <- AverageGenes
248 
+}
249 
+return(PEAmatrix)
250 
+}
251 
+
252 
+
253 
+
254 
+
255 
+
256 
+
257 
+
258 
+
259 
+
260 
+
261 
+
262 
+
263 
+#' @title For TCGA data get human pathway data and creates a measure of cross-talk among pathways
264 
+#' @description euc_dist_crtlk creates a matrix with euclidean distance for pairwise pathways
265 
+#' @param dataFilt TCGA matrix
266 
+#' @param pathway pathway data
267 
+#' @export
268 
+#' @return a matrix value for each pathway
269 
+#' @examples
270 
+#' score_euc_dista<-euc_dist_crtlk(dataFilt=tumo[,1:2],path)
271 
+euc_dist_crtlk <- function(dataFilt,pathway){
272 
+  PEAmatrix<-average(dataFilt,pathway)
273 
+  #step 5 distance
274 
+  # EUCLIDEA DISTANCE
275 
+  df=combn(rownames(PEAmatrix),2) # possibili relazioni tra i pathway
276 
+  df=t(df)
277 
+  ma_d<-matrix(0,nrow(df),ncol(PEAmatrix)) # creo matrix che conterr? le distanze
278 
+  colnames(ma_d)<-colnames(PEAmatrix) # colnames conterr? il nome dei pazienti
279 
+  for ( p in 1: ncol(PEAmatrix)){ # per ogni paziente
280 
+    patients <- (PEAmatrix)[,p]
281 
+    distance<-dist(patients) # calcolo distanza EUCLIDEA tra le possibile combinazioni
282 
+    ma_d[,p]<-distance
283 
+  }
284 
+  euc_dist<-cbind(df,ma_d) # inserisco label con le relazioni tra i pathway
285 
+  return(euc_dist)
286 
+}
287 
+
288 
+
289 
+
290 
+
291 
+#' @title For TCGA data get human pathway data and creates a measure of standard deviations among pathways
292 
+#' @description st_dv creates a matrix with standard deviation for pathways
293 
+#' @param DataMatrix TCGA matrix
294 
+#' @param pathway pathway data
295 
+#' @export
296 
+#' @return a matrix value for each pathway
297 
+#' @examples
298 
+#' stand_dev<-st_dv(DataMatrix=tumo[,1:2],pathway=path)
299 
+st_dv<-function(DataMatrix,pathway){
300 
+#DataMatrix<-dataFilt
301 
+
302 
+#dataFilt[ , "new.col"] <- gsub("\\|.*", "", rownames(dataFilt))
303 
+#DataMatrix<-dataFilt[which(dataFilt$new.col!="?"),]
304 
+#DataMatrix <- subset(DataMatrix, !duplicated(DataMatrix$new.col))
305 
+#rownames(DataMatrix)<-DataMatrix$new.col
306 
+#DataMatrix$new.col<-NULL
307 
+
308 
+PEAmatrix_sd <- matrix( 0, ncol(pathway),ncol(DataMatrix))
309 
+rownames(PEAmatrix_sd) <- colnames(pathway)
310 
+colnames(PEAmatrix_sd) <-  colnames(DataMatrix)
311 
+for ( k in 1: nrow(PEAmatrix_sd)){
312 
+  print(colnames(pathway)[k])
313 
+  currentPathway <- colnames(pathway)[k]
314 
+  currentPathway_genes_list_common <- intersect( rownames(DataMatrix), currentPathway_genes<-pathway[,k])
315 
+  currentPathway_genes_list_commonMatrix <- DataMatrix[currentPathway_genes_list_common,]
316 
+  stdev<-apply(currentPathway_genes_list_commonMatrix,2,sd) #deviazione standard dei pathway
317 
+  PEAmatrix_sd[k,] <- stdev
318 
+  }
319 
+return(PEAmatrix_sd)
320 
+}
321 
+
322 
+
323 
+
324 
+
325 
+
326 
+
327 
+#' @title For TCGA data get human pathway data and creates a measure of discriminating score among pathways
328 
+#' @description ds_score_crtlk creates a matrix with  discriminating score for pathways
329 
+#' @param dataFilt TCGA matrix
330 
+#' @param pathway pathway data
331 
+#' @export
332 
+#' @return a matrix value for each pathway
333 
+#' @examples
334 
+#' cross_talk_st_dv<-ds_score_crtlk(dataFilt=tumo[,1:2],pathway=path)
335 
+ds_score_crtlk<-function(dataFilt,pathway){
336 
+  PEAmatrix<-average(dataFilt,pathway)
337 
+  #step 5 distance
338 
+  # EUCLIDEA DISTANCE
339 
+  df=combn(rownames(PEAmatrix),2) # possibili relazioni tra i pathway
340 
+  df=t(df)
341 
+  ma_d<-matrix(0,nrow(df),ncol(PEAmatrix)) # creo matrix che conterr? le distanze
342 
+  colnames(ma_d)<-colnames(PEAmatrix) # colnames conterr? il nome dei pazienti
343 
+  for ( p in 1: ncol(PEAmatrix)){ # per ogni paziente
344 
+    patients <- (PEAmatrix)[,p]
345 
+    distance<-dist(patients) # calcolo distanza EUCLIDEA tra le possibile combinazioni
346 
+    ma_d[,p]<-distance
347 
+  }
348 
+  PEAmatrix_sd<-st_dv(dataFilt,pathway)
349 
+  df=combn(rownames(PEAmatrix_sd),2)
350 
+  df=t(df)
351 
+  ma<-matrix(0,nrow(df),ncol(PEAmatrix_sd)) # creo matrix che conterr? le somme delle dev st
352 
+  colnames(ma)<-colnames(PEAmatrix_sd) # colnames conterr? il nome dei pazienti
353 
+  for ( p in 1: ncol(PEAmatrix_sd)){ # per ogni paziente
354 
+    patients <- (PEAmatrix_sd)[,p]
355 
+    out <- apply(df, 1, function(x) sum(patients[x])) # calcolo somma delle dev standard tra le possibili combinazioni
356 
+    ma[,p]<-out
357 
+  }
358 
+  score<-ma_d/ma # discriminating score M1-M2/S1+S2
359 
+  score<- cbind(df,score)
360 
+return(score)
361 
+}
362 
+
363 
+
364 
+
365 
+#' @title SVM classification for each feature
366 
+#' @description svm class creates a list with auc value
367 
+#' @param TCGA_matrix gene expression matrix
368 
+#' @param nfs nfs split data into a training  and test set
369 
+#' @param tumour barcode samples for a class
370 
+#' @param normal barcode samples for another class
371 
+#' @export
372 
+#' @importFrom e1071 tune svm
373 
+#' @importFrom ROCR prediction performance
374 
+#' @importFrom  grDevices rainbow
375 
+#' @return a list with AUC value for pairwise pathway
376 
+#' @examples
377 
+#' nf <- 60
378 
+#' res_class<-svm_classification(TCGA_matrix=score_euc_dist,nfs=nf,
379 
+#' normal=colnames(norm[,1:10]),tumour=colnames(tumo[,1:10]))
380 
+svm_classification<-function(TCGA_matrix,tumour,normal,nfs){
381 
+  #library("e1071")
382 
+  #library(ROCR)
383 
+
384 
+  scoreMatrix <- as.data.frame(TCGA_matrix[,3:ncol(TCGA_matrix)])
385 
+  scoreMatrix <-as.data.frame(scoreMatrix)
386 
+  for( i in 1: ncol(scoreMatrix)){
387 
+    scoreMatrix[,i] <- as.numeric(as.character(scoreMatrix[,i]))
388 
+  }
389 
+
390 
+  TCGA_matrix[,1] <- gsub(" ", "_", TCGA_matrix[,1])
391 
+  d<-sub('_-_Homo_sapiens_*', '', TCGA_matrix[,1])
392 
+  #d_pr<-sub(')*', '', DataMatrix[,1])
393 
+
394 
+  d_pr<- gsub("(human)", "", d, fixed="TRUE")
395 
+  d_pr <- gsub("_", "", d_pr)
396 
+  d_pr <- gsub("-", "", d_pr)
397 
+
398 
+  TCGA_matrix[,2] <- gsub(" ", "_", TCGA_matrix[,2])
399 
+  d2<-sub('_-_Homo_sapiens_(human)*', '', TCGA_matrix[,2])
400 
+  d_pr2<- gsub("(human)", "", d2, fixed="TRUE")
401 
+  d_pr2 <- gsub("_", "", d_pr2)
402 
+  d_pr2 <- gsub("-", "", d_pr2)
403 
+
404 
+  PathwaysPair <- paste( as.matrix(d_pr), as.matrix(d_pr2),sep="_" )
405 
+
406 
+  rownames(scoreMatrix) <-PathwaysPair
407 
+
408 
+
409 
+  tDataMatrix<-as.data.frame(t(scoreMatrix))
410 
+  #tDataMatrix$Target[,1]<-0
411 
+
412 
+  tDataMatrix<-cbind(Target=0,tDataMatrix )
413 
+
414 
+  tum<-intersect(rownames(tDataMatrix),tumour)
415 
+  nor<-intersect(rownames(tDataMatrix),normal)
416 
+  #tDataMatrix$
417 
+
418 
+  Dataset_g1<-tDataMatrix[nor,]
419 
+  Dataset_g3<- tDataMatrix[tum,]
420 
+
421 
+
422 
+#training=read.table('C:/Users/UserInLab05/Desktop/trai.txt',header = TRUE)
423 
+#testset=read.table('C:/Users/UserInLab05/Desktop/test.txt',header = TRUE)
424 
+
425 
+  Dataset_g1$Target <- 0
426 
+  Dataset_g3$Target<-1
427 
+#Dataset_g3 <- Dataset_g3[Dataset_g3$Target <- 1, ]
428 
+
429 
+tab_g1_training <- sample(rownames(Dataset_g1),round(nrow(Dataset_g1) / 100 * nfs ))
430 
+tab_g3_training <- sample(rownames(Dataset_g3),round(nrow(Dataset_g3) / 100 * nfs ))
431 
+tab_g1_testing <- setdiff(rownames(Dataset_g1),tab_g1_training)
432 
+tab_g3_testing <- setdiff(rownames(Dataset_g3),tab_g3_training)
433 
+
434 
+FR<-intersect(rownames(Dataset_g1),tab_g1_training)
435 
+
436 
+#rownames(Dataset_g1)<-Dataset_g1[,1]
437 
+G1<-Dataset_g1[FR,]
438 
+
439 
+FR1<-intersect(rownames(Dataset_g3),tab_g3_training)
440 
+#rownames(Dataset_g3)<-Dataset_g3$ID
441 
+
442 
+G3<-Dataset_g3[FR1,]
443 
+training<-rbind(G1,G3)
444 
+
445 
+inter1<-intersect(rownames(Dataset_g1),tab_g1_testing)
446 
+#rownames(Dataset_g1)<-Dataset_g1$ID
447 
+
448 
+G1_testing<-Dataset_g1[inter1,]
449 
+
450 
+inter2<-intersect(rownames(Dataset_g3),tab_g3_testing)
451 
+#rownames(Dataset_g3)<-Dataset_g3$ID
452 
+G3_testing<-Dataset_g3[inter2,]
453 
+
454 
+testing<-rbind(G1_testing,G3_testing)
455 
+
456 
+x <- subset(training, select=-Target)
457 
+y <- training$Target
458 
+#testing[,2]<-NULL
459 
+z<-subset(testing, select=-Target)
460 
+
461 
+zi<-testing$Target
462 
+
463 
+auc.df<-list()
464 
+svm_model_after_tune_COMPL<-list()
465 
+for( k in 2: ncol(training)){
466 
+  print(colnames(training)[k])
467 
+  svm_tune <- tune(svm, train.x=x, train.y=y,
468 
+                   kernel="radial", ranges=list(cost=10^(-1:2), gamma=c(.5,1,2)),cross=10)
469 
+  #print(svm_tune)
470 
+
471 
+  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)
472 
+
473 
+
474 
+  #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)
475 
+  #summary(svm_model_after_tune)
476 
+
477 
+  j=k-1
478 
+  z2=z[,j]
479 
+  z3<-as.data.frame(z2)
480 
+  #rownames(z3)<-rownames(z)
481 
+  #colnames(z3)<-as.character(paste("X",j,sep = ""))
482 
+  colnames(z3)<-colnames(z)[j]
483