-
Notifications
You must be signed in to change notification settings - Fork 0
/
lihc.py
154 lines (118 loc) · 8.22 KB
/
lihc.py
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
66
67
68
69
70
71
72
73
74
75
76
77
78
79
80
81
82
83
84
85
86
87
88
89
90
91
92
93
94
95
96
97
98
99
100
101
102
103
104
105
106
107
108
109
110
111
112
113
114
115
116
117
118
119
120
121
122
123
124
125
126
127
128
129
130
131
132
133
134
135
136
137
138
139
140
141
142
143
144
145
146
147
148
149
150
151
152
153
154
import os
import rpy2.robjects as robjects
import json
def analysisLIHC(idxAnalysis, file_miRNA, file_RNA, pvalue, foldchange, contrasts, geneSelected):
root = "datasetTCGA" + os.sep + "LIHC" + os.sep
path_file_miRNA = root + "miRNASeq" + os.sep + file_miRNA
path_file_RNA = root + "RNASeq" + os.sep + file_RNA
name_file_rdata = "static/tempAnalysis" + os.sep + "LIHC_workspace.RData"
name_file_json = "static/tempAnalysis" + os.sep + "LIHC_" + pvalue + "_" + foldchange + "_" + contrasts + ".json"
name_file_heatmap = "static/tempAnalysis" + os.sep + "LIHC_" + pvalue + "_" + foldchange + "_" + contrasts + "_heatmap.jpg"
name_file_boxplot = "static/tempAnalysis" + os.sep + "LIHC_" + pvalue + "_" + foldchange + "_" + contrasts + "_" + geneSelected + "_boxplot.jpg"
name_file_mithril_1 = "static/tempAnalysis" + os.sep + "input_mithril_LIHC_" + contrasts
name_file_mithril_2 = "static/tempAnalysis" + os.sep + "input_mithril_LIHC_" + contrasts + ".txt" #questo è per i contrasti che non sono ALL
robjects.r('''
library("limma")
library("Biobase")
library("gplots")
library("data.table")
library("jsonlite")
if({0} == 4) {{ #Creazione File di input per mithril
load(file="{3}")
if({4} == 0) {{
for (i in 1:5) {{
name.file <- paste(paste("{11}", i, sep="_"), ".txt", sep="")
write.table(results[i], name.file, sep="\t" ,row.names=TRUE, col.names = FALSE, quote = FALSE)
}}
}} else {{
write.table(results[1], "{12}", sep="\t" ,row.names=TRUE, col.names = FALSE, quote = FALSE)
}}
}} else if({0} == 3) {{ #Creazione boxplot per singolo gene
load(file="{3}")
gene <- "{9}"
exp.values <- as.vector(normalized.expressions$E[gene, rownames(df.patient)])
boxplot.data <- data.frame(Sample.Category=df.patient$ajcc_pathologic_tumor_stage, Sample.Value=exp.values, row.names=rownames(df.patient))
jpeg(file="{10}")
boxplot(Sample.Value~Sample.Category, data=boxplot.data, main=paste0("Evaluation of ", gene), xlab="Sample Category", ylab="Expression Value")
dev.off()
}} else {{
if({0} == 1) {{ #Analisi per l'estrazione dei biomarcatori
# =========== LOAD BIOSPECIMEN CLINICAL ===========
col.df.1 <- c("bcr_patient_uuid", "bcr_patient_barcode", "tumor_status", "ajcc_pathologic_tumor_stage", "histologic_diagnosis", "gender", "vital_status")
df.patient <- read.table("datasetTCGA/LIHC/BiospecimenClinicalData/nationwidechildrens.org_clinical_patient_lihc.txt", header = T, sep = "\t", quote="")[col.df.1]
df.patient <- df.patient[-c(1, 2), ]
# =========== LOAD miRNA Seq ===========
df.miRNA <- read.table("{1}", header = T, sep = "\t", check.names = FALSE, row.names = 1)
df.miRNA <- data.matrix(df.miRNA[-1, seq(from = 2, to = ncol(df.miRNA), by = 2)])
# =========== LOAD RNA Seq ===========
df.RNA <- read.table("{2}", header = T, sep = "\t", check.names = FALSE, row.names = 1)
df.RNA <- data.matrix(df.RNA)
# =========== INTEGRATION DATA ===========
rownames(df.patient) <- as.vector(df.patient$bcr_patient_barcode)
df.patient <- df.patient[-which(
df.patient$ajcc_pathologic_tumor_stage == "[Not Available]" |
df.patient$ajcc_pathologic_tumor_stage == "[Discrepancy]" |
df.patient$ajcc_pathologic_tumor_stage == ""
),] #Rimuovo tutti i levels '[Not Available], [Discrepancy], vuoti'
df.patient$ajcc_pathologic_tumor_stage <- droplevels(df.patient)$ajcc_pathologic_tumor_stage #rimuovo i livelli non utilizzati
col.patient.barcode <- as.vector(unique(df.patient$bcr_patient_barcode))
idx.col.miRNA <- array(unlist(lapply(col.patient.barcode, function(x) which(colnames(df.miRNA) %like% as.character(x)))))
idx.col.RNA <- array(unlist(lapply(col.patient.barcode, function(x) which(colnames(df.RNA) %like% as.character(x)))))
df.miRNA <- as.data.frame(df.miRNA[, idx.col.miRNA])
names(df.miRNA) <- substr(names(df.miRNA),-1, 12)
df.RNA <- as.data.frame(df.RNA[, idx.col.RNA])
names(df.RNA) <- substr(names(df.RNA),-1, 12)
col.union <- intersect(colnames(df.RNA), colnames(df.miRNA))
df.union <- data.matrix(rbind(df.miRNA[, col.union], df.RNA[, col.union]))
df.patient <- df.patient[match(colnames(df.union), df.patient$bcr_patient_barcode),]
design.original <- model.matrix(~0+ajcc_pathologic_tumor_stage, data=df.patient)
colnames(design.original) <- c("S1", "S2", "S3", "S3a", "S3b", "S3c", "S4", "S4a", "S4b")
# =========== ANALYSIS ===========
normalized.expressions <- voom(df.union, design.original)
initial.model <- lmFit(normalized.expressions, design.original)
limma.model <- eBayes(initial.model)
#ogni volta usate lo stadio precedente come controllo per vedere quali geni differiscono
contrasts <-makeContrasts(S2-S1, S3-S2, S3a-S3, S3b-S3a, S3c-S3b, S4-S3c, S4a-S4, S4b-S4a, levels=design.original)
contrasts.model <- eBayes(contrasts.fit(limma.model, contrasts))
}} else if({0} == 2) {{ #Analisi per l'estrazione dei biomarcatori (Custom)
load(file="{3}")
}}
if({4} == 0) {{
results <- topTableF(contrasts.model, number=nrow(df.union), adjust.method="BH", p.value={5}, lfc={6})
}} else {{
results <- topTable(contrasts.model, coef={4}, number=nrow(df.union), adjust.method="BH", p.value={5}, lfc={6})
}}
#print(nrow(results))
if(is.data.frame(results) && nrow(results)!=0) {{
json <- toJSON(results, pretty = T)
#cat(x) #serve a stamparlo
write(json, "{7}")
if(nrow(results) > 1) {{ #evito di stampare la heatmap, se il dataframe result è composto di una sola riga
de.genes <- rownames(results)
#de.expressions <- normalized.expressions$E[de.genes,]
de.expressions <- matrix(as.numeric(unlist(normalized.expressions$E[de.genes,])), nrow=nrow(normalized.expressions$E[de.genes,]))
jpeg(file="{8}")
heatmap.2(de.expressions, col=redgreen(100), scale="row", key=TRUE, symkey=FALSE, density.info="none", trace="none")
dev.off()
}}
}}
save.image(file="{3}")
}}
'''.format(idxAnalysis, path_file_miRNA, path_file_RNA, name_file_rdata, contrasts, pvalue,
foldchange, name_file_json, name_file_heatmap, geneSelected, name_file_boxplot, name_file_mithril_1, name_file_mithril_2))
nGenes = 0
if os.path.isfile(name_file_json) and (idxAnalysis == "1" or idxAnalysis == "2"):
#apro il json appena creato
with open(name_file_json, 'r') as f:
datastore = json.load(f)
htmlGenes = '<option value="">Seleziona...</option>'
htmlAllGenes = ''
for idx,value in enumerate(datastore):
htmlGenes += '<option value="'+value['_row']+'">'+value['_row']+'</option>'
htmlAllGenes += value['_row']+"<br />"
nGenes += 1
# questo html, mi serve per inserirlo nella lista dei geni
return htmlGenes, htmlAllGenes, nGenes, 0
elif idxAnalysis == "3" or idxAnalysis == "4": # è importante questa condizione, perchè allora causerà un errore sulla creazione del boxplot o nell'estrazione dei file di input
return '', '', nGenes, 0
return '', '', nGenes, 1