comparison annotationRmn2DGlobale.R @ 3:546c7ccd2ed4 draft default tip

"planemo upload for repository https://github.com/workflow4metabolomics/tools-metabolomics commit 911f4beba3dcb25c1033e8239426f8f763683523"
author workflow4metabolomics
date Fri, 04 Feb 2022 09:01:11 +0000
parents dff7bde22102
children
comparison
equal deleted inserted replaced
2:dff7bde22102 3:546c7ccd2ed4
1 ########################################################################################################################################### 1 ###################################################################################################
2 # ANNOTATION SPECTRE 2D MATRICE COMPLEXE BASEE SUR UNE (OU PLUSIEURS) SEQUENCE(s) RMN # 2 # ANNOTATION SPECTRE 2D MATRICE COMPLEXE BASEE SUR UNE (OU PLUSIEURS) SEQUENCE(s) #
3 # template : dataframe contenant la liste des couples de deplacements chimiques de la matrice complexe a annoter # 3 # template : dataframe contenant la liste des couples de deplacements chimiques de la matrice complexe a annoter #
4 # cosy : 1 si sequence a utiliser / 0 sinon # 4 # cosy : 1 si sequence a utiliser / 0 sinon #
5 # hmbc : 1 si sequence a utiliser / 0 sinon # 5 # hmbc : 1 si sequence a utiliser / 0 sinon #
6 # hsqc : 1 si sequence a utiliser / 0 sinon # 6 # hsqc : 1 si sequence a utiliser / 0 sinon #
7 # jres : 1 si sequence a utiliser / 0 sinon # 7 # jres : 1 si sequence a utiliser / 0 sinon #
8 # tocsy : 1 si sequence a utiliser / 0 sinon # 8 # tocsy : 1 si sequence a utiliser / 0 sinon #
9 # tolPpm1 : tolerance autorisee autour de la valeur1 du couple de deplacements chimiques # 9 # tolPpm1 : tolerance autorisee autour de la valeur1 du couple de deplacements chimiques #
10 # tolPpm2HJRes : tolerance autorisee autour de la valeur2 du couple de deplacements chimiques si H dans dimension 2 # 10 # tolPpm2HJRes : tolerance autorisee autour de la valeur2 du couple de deplacements chimiques si H dans dimension 2 #
11 # tolPpm2C : tolerance autorisee autour de la valeur2 du couple de deplacements chimiques si C dans dimension 2 # 11 # tolPpm2C : tolerance autorisee autour de la valeur2 du couple de deplacements chimiques si C dans dimension 2 #
12 # seuil : valeur du score de presence en deça de laquelle les metabolites annotes ne sont pas retenus # 12 # seuil : valeur du score de presence en dela de laquelle les metabolites annotes ne sont pas retenus #
13 # unicite : boolean pour ne retenir que les ... # 13 # unicite : boolean pour ne retenir que les ... #
14 ########################################################################################################################################### 14 ###################################################################################################
15 ## CALCUL MOYENNE SANS VALEUR(S) MANQUANTE(S) 15 ## CALCUL MOYENNE SANS VALEUR(S) MANQUANTE(S)
16 mean.rmNa <- function(x) 16 mean.rmNa <- function(x) {
17 { 17 mean(x, na.rm = TRUE)
18 mean(x, na.rm=TRUE)
19 } 18 }
20 19
21 annotationRmn2DGlobale <- function(template, tolPpm1=0.01, tolPpm2HJRes=0.002, tolPpm2C=0.5, cosy=1, hmbc=1, hsqc=1, jres=1, tocsy=1, 20 annotationRmn2DGlobale <- function(template, tolPpm1 = 0.01, tolPpm2HJRes = 0.002, tolPpm2C = 0.5, cosy = 1, hmbc = 1, hsqc = 1, jres = 1, tocsy = 1, seuil, unicite = "NO") {
22 seuil, unicite="NO")
23 {
24 ## Initialisation 21 ## Initialisation
25 options (max.print=999999999) 22 options(max.print = 999999999)
26 annotationCOSY <- data.frame() 23 annotationCOSY <- data.frame()
27 annotationHMBC <- data.frame() 24 annotationHMBC <- data.frame()
28 annotationHSQC <- data.frame() 25 annotationHSQC <- data.frame()
29 annotationJRES <- data.frame() 26 annotationJRES <- data.frame()
30 annotationTOCSY <- data.frame() 27 annotationTOCSY <- data.frame()
32 dataCOSY <- "NA" 29 dataCOSY <- "NA"
33 dataHMBC <- "NA" 30 dataHMBC <- "NA"
34 dataHSQC <- "NA" 31 dataHSQC <- "NA"
35 dataJRES <- "NA" 32 dataJRES <- "NA"
36 dataTOCSY <- "NA" 33 dataTOCSY <- "NA"
37 34
38 ## Application seuil seulement si annotation avec 1 seule sequence 35 ## Application seuil seulement si annotation avec 1 seule sequence
39 ## seuilPls2D <- 0
40 ## if ((sum(cosy, hmbc, hsqc, jres, tocsy)) == 1)
41 ## seuilPls2D <- seuil
42 seuilPls2D <- seuil 36 seuilPls2D <- seuil
43 37
44 if (cosy == 1) 38 if (cosy == 1) {
45 { 39 matrice.cosy <- read.xlsx(template, sheet = "COSY", startRow = 2, colNames = TRUE, rowNames = FALSE, cols = 1:3, na.strings = "NA")
46 matrice.cosy <- read.xlsx(template, sheet="COSY", startRow=2, colNames=TRUE, rowNames=FALSE, cols=1:3, na.strings="NA")
47 matrice.cosy <- matrice.cosy[matrice.cosy$peak.index != "x", ] 40 matrice.cosy <- matrice.cosy[matrice.cosy$peak.index != "x", ]
48 annotationCOSY <- annotationRmn2D(matrice.cosy, BdDReference_COSY, "COSY", ppm1Tol=tolPpm1, ppm2Tol=tolPpm1, seuil=seuilPls2D, 41 annotationCOSY <- annotationRmn2D(matrice.cosy, BdDReference_COSY, "COSY", ppm1Tol = tolPpm1, ppm2Tol = tolPpm1, seuil = seuilPls2D, unicite = unicite)
49 unicite=unicite) 42 dataCOSY <- data.frame(Metabolite = str_to_lower(annotationCOSY$liste_resultat$Metabolite), score.COSY = annotationCOSY$liste_resultat$score)
50 dataCOSY <- data.frame(Metabolite=str_to_lower(annotationCOSY$liste_resultat$Metabolite), score.COSY=annotationCOSY$liste_resultat$score)
51 dataCOSY <- unique.data.frame(dataCOSY) 43 dataCOSY <- unique.data.frame(dataCOSY)
52 } 44 }
53 45
54 if (hmbc == 1) 46 if (hmbc == 1) {
55 { 47 matrice.hmbc <- read.xlsx(template, sheet = "HMBC", startRow = 2, colNames = TRUE, rowNames = FALSE, cols = 1:3, na.strings = "NA")
56 matrice.hmbc <- read.xlsx(template, sheet="HMBC", startRow=2, colNames=TRUE, rowNames=FALSE, cols=1:3, na.strings="NA")
57 matrice.hmbc <- matrice.hmbc[matrice.hmbc$peak.index != "x", ] 48 matrice.hmbc <- matrice.hmbc[matrice.hmbc$peak.index != "x", ]
58 annotationHMBC <- annotationRmn2D(matrice.hmbc, BdDReference_HMBC, "HMBC", ppm1Tol=tolPpm1, ppm2Tol=tolPpm2C, seuil=seuilPls2D, 49 annotationHMBC <- annotationRmn2D(matrice.hmbc, BdDReference_HMBC, "HMBC", ppm1Tol = tolPpm1, ppm2Tol = tolPpm2C, seuil = seuilPls2D, unicite = unicite)
59 unicite=unicite) 50 dataHMBC <- data.frame(Metabolite = str_to_lower(annotationHMBC$liste_resultat$Metabolite), score.HMBC = annotationHMBC$liste_resultat$score)
60 dataHMBC <- data.frame(Metabolite=str_to_lower(annotationHMBC$liste_resultat$Metabolite), score.HMBC=annotationHMBC$liste_resultat$score)
61 dataHMBC <- unique.data.frame(dataHMBC) 51 dataHMBC <- unique.data.frame(dataHMBC)
62 } 52 }
63 53
64 if (hsqc == 1) 54 if (hsqc == 1) {
65 { 55 matrice.hsqc <- read.xlsx(template, sheet = "HSQC", startRow = 2, colNames = TRUE, rowNames = FALSE, cols = 1:3, na.strings = "NA")
66 matrice.hsqc <- read.xlsx(template, sheet="HSQC", startRow=2, colNames=TRUE, rowNames=FALSE, cols=1:3, na.strings="NA")
67 matrice.hsqc <- matrice.hsqc[matrice.hsqc$peak.index != "x", ] 56 matrice.hsqc <- matrice.hsqc[matrice.hsqc$peak.index != "x", ]
68 annotationHSQC <- annotationRmn2D(matrice.hsqc, BdDReference_HSQC, "HSQC", ppm1Tol=tolPpm1, ppm2Tol=tolPpm2C, seuil=seuilPls2D, 57 annotationHSQC <- annotationRmn2D(matrice.hsqc, BdDReference_HSQC, "HSQC", ppm1Tol = tolPpm1, ppm2Tol = tolPpm2C, seuil = seuilPls2D, unicite = unicite)
69 unicite=unicite) 58 dataHSQC <- data.frame(Metabolite = str_to_lower(annotationHSQC$liste_resultat$Metabolite), score.HSQC = annotationHSQC$liste_resultat$score)
70 dataHSQC <- data.frame(Metabolite=str_to_lower(annotationHSQC$liste_resultat$Metabolite), score.HSQC=annotationHSQC$liste_resultat$score)
71 dataHSQC <- unique.data.frame(dataHSQC) 59 dataHSQC <- unique.data.frame(dataHSQC)
72 } 60 }
73 61
74 if (jres == 1) 62 if (jres == 1) {
75 { 63 matrice.jres <- read.xlsx(template, sheet = "JRES", startRow = 2, colNames = TRUE, rowNames = FALSE, cols = 1:3, na.strings = "NA")
76 matrice.jres <- read.xlsx(template, sheet="JRES", startRow=2, colNames=TRUE, rowNames=FALSE, cols=1:3, na.strings="NA")
77 matrice.jres <- matrice.jres[matrice.jres$peak.index != "x", ] 64 matrice.jres <- matrice.jres[matrice.jres$peak.index != "x", ]
78 annotationJRES <- annotationRmn2D(matrice.jres, BdDReference_JRES, "JRES", ppm1Tol=tolPpm1, ppm2Tol=tolPpm2HJRes, seuil=seuilPls2D, 65 annotationJRES <- annotationRmn2D(matrice.jres, BdDReference_JRES, "JRES", ppm1Tol = tolPpm1, ppm2Tol = tolPpm2HJRes, seuil = seuilPls2D, unicite = unicite)
79 unicite=unicite) 66 dataJRES <- data.frame(Metabolite = str_to_lower(annotationJRES$liste_resultat$Metabolite), score.JRES = annotationJRES$liste_resultat$score)
80 dataJRES <- data.frame(Metabolite=str_to_lower(annotationJRES$liste_resultat$Metabolite), score.JRES=annotationJRES$liste_resultat$score)
81 dataJRES <- unique.data.frame(dataJRES) 67 dataJRES <- unique.data.frame(dataJRES)
82 } 68 }
83 69
84 if (tocsy == 1) 70 if (tocsy == 1) {
85 { 71 matrice.tocsy <- read.xlsx(template, sheet = "TOCSY", startRow = 2, colNames = TRUE, rowNames = FALSE, cols = 1:3, na.strings = "NA")
86 matrice.tocsy <- read.xlsx(template, sheet="TOCSY", startRow=2, colNames=TRUE, rowNames=FALSE, cols=1:3, na.strings="NA")
87 matrice.tocsy <- matrice.tocsy[matrice.tocsy$peak.index != "x", ] 72 matrice.tocsy <- matrice.tocsy[matrice.tocsy$peak.index != "x", ]
88 annotationTOCSY <- annotationRmn2D(matrice.tocsy, BdDReference_TOCSY, "TOCSY", ppm1Tol=tolPpm1, ppm2Tol=tolPpm1, seuil=seuilPls2D, 73 annotationTOCSY <- annotationRmn2D(matrice.tocsy, BdDReference_TOCSY, "TOCSY", ppm1Tol = tolPpm1, ppm2Tol = tolPpm1, seuil = seuilPls2D, unicite = unicite)
89 unicite=unicite) 74 dataTOCSY <- data.frame(Metabolite = str_to_lower(annotationTOCSY$liste_resultat$Metabolite), score.TOCSY = annotationTOCSY$liste_resultat$score)
90 dataTOCSY <- data.frame(Metabolite=str_to_lower(annotationTOCSY$liste_resultat$Metabolite), score.TOCSY=annotationTOCSY$liste_resultat$score)
91 dataTOCSY <- unique.data.frame(dataTOCSY) 75 dataTOCSY <- unique.data.frame(dataTOCSY)
92 } 76 }
93 77
94 sequencesCombinationAverageScoreSeuil <- data.frame() 78 seqCombiMeanScoreSeuil <- data.frame()
95 sequencesCombinationAverageScoreSeuilFiltre <- data.frame() 79 seqCombiMeanScoreSeuilFiltre <- data.frame()
96 80
97 ## CONCATENATION RESULTATS DIFFERENTES SEQUENCES 81 ## CONCATENATION RESULTATS DIFFERENTES SEQUENCES
98 data2D <- list(dataCOSY, dataHMBC, dataHSQC, dataJRES, dataTOCSY) 82 data2D <- list(dataCOSY, dataHMBC, dataHSQC, dataJRES, dataTOCSY)
99 whichSequenceNaN <- which((data2D != "NA")) 83 whichSequenceNaN <- which((data2D != "NA"))
100 data2D <- data2D[whichSequenceNaN] 84 data2D <- data2D[whichSequenceNaN]
101 sequencesCombination <- data.frame(data2D[1]) 85 sequencesCombination <- data.frame(data2D[1])
102 sequencesCombinationAverageScore <- sequencesCombination 86 seqCombiMeanScore <- sequencesCombination
103 87
104 ## Si une seule sequence et seuil sur score = filtre applique dans la fonction annotationRmn2D 88 ## Si une seule sequence et seuil sur score = filtre applique dans la fonction annotationRmn2D
105 if (length(data2D) >= 2) 89 if (length(data2D) >= 2) {
106 {
107 ## CONCATENATION SCORE PAR SEQUENCE 90 ## CONCATENATION SCORE PAR SEQUENCE
108 for (l in 2:length(data2D)) 91 for (l in 2:length(data2D))
109 sequencesCombination <- merge.data.frame(sequencesCombination, data2D[l], by="Metabolite", all.x=TRUE, all.y=TRUE) 92 sequencesCombination <- merge.data.frame(sequencesCombination, data2D[l], by = "Metabolite", all.x = TRUE, all.y = TRUE)
110 93
111 ## SCORE MOYEN (sans prise en compte valeurs manquantes) 94 ## Replacement of NA values due to mis annotation
112 meanScore <- apply(sequencesCombination[, -1], 1, FUN=mean.rmNa) 95 for (m in seq_len(nrow(sequencesCombination))) {
113 sequencesCombinationAverageScore <- cbind.data.frame(sequencesCombination, averageScore=meanScore) 96 COSYcompound <- sort(names(BdDReference_COSY))
114 ## SUPPRESSION METABOLITE AVEC SCORE MOYEN < SEUIL 97 HMBCcompound <- sort(names(BdDReference_HMBC))
115 ## sequencesCombinationAverageScoreSeuilFiltre <- filter(sequencesCombinationAverageScore, averageScore >= seuil) 98 HSQCcompound <- sort(names(BdDReference_HSQC))
116 sequencesCombinationAverageScoreSeuilFiltre <- sequencesCombinationAverageScore[sequencesCombinationAverageScore$averageScore > seuil, ] 99 JREScompound <- sort(names(BdDReference_JRES))
100 TOCSYcompound <- sort(names(BdDReference_TOCSY))
101
102 if (is.na(sequencesCombination[m, 2])) {
103 compound <- as.character(sequencesCombination[m, 1])
104 for (c in seq_len(length(COSYcompound)))
105 if (str_to_lower(compound) == str_to_lower(COSYcompound[c]))
106 sequencesCombination[m, 2] <- 0
107 }
108
109 if (is.na(sequencesCombination[m, 3])) {
110 compound <- as.character(sequencesCombination[m, 1])
111 for (c in seq_len(length(HMBCcompound)))
112 if (str_to_lower(compound) == str_to_lower(HMBCcompound[c]))
113 sequencesCombination[m, 3] <- 0
114 }
115
116 if (is.na(sequencesCombination[m, 4])) {
117 compound <- as.character(sequencesCombination[m, 1])
118 for (c in seq_len(length(HSQCcompound)))
119 if (str_to_lower(compound) == str_to_lower(HSQCcompound[c]))
120 sequencesCombination[m, 4] <- 0
121 }
122
123 if (is.na(sequencesCombination[m, 5])) {
124 compound <- as.character(sequencesCombination[m, 1])
125 for (c in seq_len(length(JREScompound)))
126 if (str_to_lower(compound) == str_to_lower(JREScompound[c]))
127 sequencesCombination[m, 5] <- 0
128 }
129
130 if (is.na(sequencesCombination[m, 6])) {
131 compound <- as.character(sequencesCombination[m, 1])
132 for (c in seq_len(length(TOCSYcompound)))
133 if (str_to_lower(compound) == str_to_lower(TOCSYcompound[c]))
134 sequencesCombination[m, 6] <- 0
135 }
117 } 136 }
118 137
119 return(list(COSY=annotationCOSY, HMBC=annotationHMBC, HSQC=annotationHSQC, JRES=annotationJRES, TOCSY=annotationTOCSY, 138 ## SCORE MOYEN (sans prise en compte valeurs manquantes)
120 combination=sequencesCombinationAverageScoreSeuilFiltre)) 139 meanScore <- round(apply(sequencesCombination[, -1], 1, FUN = mean.rmNa), 2)
140 seqCombiMeanScore <- cbind.data.frame(sequencesCombination, averageScore = meanScore)
141
142 ## SUPPRESSION METABOLITE AVEC SCORE MOYEN < SEUIL
143 seqCombiMeanScoreSeuilFiltre <- seqCombiMeanScore[seqCombiMeanScore$averageScore > seuil, ]
144 }
145
146 return(list(COSY = annotationCOSY, HMBC = annotationHMBC, HSQC = annotationHSQC, JRES = annotationJRES, TOCSY = annotationTOCSY, combination = seqCombiMeanScoreSeuilFiltre))
121 } 147 }