Mercurial > repos > marie-tremblay-metatoul > 2dnmrannotation

comparison annotationRmn2DGlobale.R @ 2:dff7bde22102 draft

Find changesets by keywords (author, files, the commit message), revision number or hash, or revset expression.
"planemo upload for repository https://github.com/workflow4metabolomics/tools-metabolomics commit b3abcb650e9b38458aa0ac5f7d838811d982ff65"
author workflow4metabolomics
date Tue, 04 Feb 2020 10:59:26 -0500
parents
children 546c7ccd2ed4
comparison
equal deleted inserted replaced
1:a4d2b1926e13 2:dff7bde22102
1 ###########################################################################################################################################
2 # ANNOTATION SPECTRE 2D MATRICE COMPLEXE BASEE SUR UNE (OU PLUSIEURS) SEQUENCE(s) RMN #
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 #
5 # hmbc : 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 #
8 # tocsy : 1 si sequence a utiliser / 0 sinon #
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 #
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 #
13 # unicite : boolean pour ne retenir que les ... #
14 ###########################################################################################################################################
15 ## CALCUL MOYENNE SANS VALEUR(S) MANQUANTE(S)
16 mean.rmNa <- function(x)
17 {
18 mean(x, na.rm=TRUE)
19 }
20
21 annotationRmn2DGlobale <- function(template, tolPpm1=0.01, tolPpm2HJRes=0.002, tolPpm2C=0.5, cosy=1, hmbc=1, hsqc=1, jres=1, tocsy=1,
22 seuil, unicite="NO")
23 {
24 ## Initialisation
25 options (max.print=999999999)
26 annotationCOSY <- data.frame()
27 annotationHMBC <- data.frame()
28 annotationHSQC <- data.frame()
29 annotationJRES <- data.frame()
30 annotationTOCSY <- data.frame()
31
32 dataCOSY <- "NA"
33 dataHMBC <- "NA"
34 dataHSQC <- "NA"
35 dataJRES <- "NA"
36 dataTOCSY <- "NA"
37
38 ## 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
43
44 if (cosy == 1)
45 {
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$peak.index != "x", ]
48 annotationCOSY <- annotationRmn2D(matrice.cosy, BdDReference_COSY, "COSY", ppm1Tol=tolPpm1, ppm2Tol=tolPpm1, seuil=seuilPls2D,
49 unicite=unicite)
50 dataCOSY <- data.frame(Metabolite=str_to_lower(annotationCOSY$liste_resultat$Metabolite), score.COSY=annotationCOSY$liste_resultat$score)
51 dataCOSY <- unique.data.frame(dataCOSY)
52 }
53
54 if (hmbc == 1)
55 {
56 matrice.hmbc <- read.xlsx(template, sheet="HMBC", startRow=2, colNames=TRUE, rowNames=FALSE, cols=1:3, na.strings="NA")
57 matrice.hmbc <- matrice.hmbc[matrice.hmbc$peak.index != "x", ]
58 annotationHMBC <- annotationRmn2D(matrice.hmbc, BdDReference_HMBC, "HMBC", ppm1Tol=tolPpm1, ppm2Tol=tolPpm2C, seuil=seuilPls2D,
59 unicite=unicite)
60 dataHMBC <- data.frame(Metabolite=str_to_lower(annotationHMBC$liste_resultat$Metabolite), score.HMBC=annotationHMBC$liste_resultat$score)
61 dataHMBC <- unique.data.frame(dataHMBC)
62 }
63
64 if (hsqc == 1)
65 {
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$peak.index != "x", ]
68 annotationHSQC <- annotationRmn2D(matrice.hsqc, BdDReference_HSQC, "HSQC", ppm1Tol=tolPpm1, ppm2Tol=tolPpm2C, seuil=seuilPls2D,
69 unicite=unicite)
70 dataHSQC <- data.frame(Metabolite=str_to_lower(annotationHSQC$liste_resultat$Metabolite), score.HSQC=annotationHSQC$liste_resultat$score)
71 dataHSQC <- unique.data.frame(dataHSQC)
72 }
73
74 if (jres == 1)
75 {
76 matrice.jres <- read.xlsx(template, sheet="JRES", startRow=2, colNames=TRUE, rowNames=FALSE, cols=1:3, na.strings="NA")
77 matrice.jres <- matrice.jres[matrice.jres$peak.index != "x", ]
78 annotationJRES <- annotationRmn2D(matrice.jres, BdDReference_JRES, "JRES", ppm1Tol=tolPpm1, ppm2Tol=tolPpm2HJRes, seuil=seuilPls2D,
79 unicite=unicite)
80 dataJRES <- data.frame(Metabolite=str_to_lower(annotationJRES$liste_resultat$Metabolite), score.JRES=annotationJRES$liste_resultat$score)
81 dataJRES <- unique.data.frame(dataJRES)
82 }
83
84 if (tocsy == 1)
85 {
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$peak.index != "x", ]
88 annotationTOCSY <- annotationRmn2D(matrice.tocsy, BdDReference_TOCSY, "TOCSY", ppm1Tol=tolPpm1, ppm2Tol=tolPpm1, seuil=seuilPls2D,
89 unicite=unicite)
90 dataTOCSY <- data.frame(Metabolite=str_to_lower(annotationTOCSY$liste_resultat$Metabolite), score.TOCSY=annotationTOCSY$liste_resultat$score)
91 dataTOCSY <- unique.data.frame(dataTOCSY)
92 }
93
94 sequencesCombinationAverageScoreSeuil <- data.frame()
95 sequencesCombinationAverageScoreSeuilFiltre <- data.frame()
96
97 ## CONCATENATION RESULTATS DIFFERENTES SEQUENCES
98 data2D <- list(dataCOSY, dataHMBC, dataHSQC, dataJRES, dataTOCSY)
99 whichSequenceNaN <- which((data2D != "NA"))
100 data2D <- data2D[whichSequenceNaN]
101 sequencesCombination <- data.frame(data2D[1])
102 sequencesCombinationAverageScore <- sequencesCombination
103
104 ## Si une seule sequence et seuil sur score = filtre applique dans la fonction annotationRmn2D
105 if (length(data2D) >= 2)
106 {
107 ## CONCATENATION SCORE PAR SEQUENCE
108 for (l in 2:length(data2D))
109 sequencesCombination <- merge.data.frame(sequencesCombination, data2D[l], by="Metabolite", all.x=TRUE, all.y=TRUE)
110
111 ## SCORE MOYEN (sans prise en compte valeurs manquantes)
112 meanScore <- apply(sequencesCombination[, -1], 1, FUN=mean.rmNa)
113 sequencesCombinationAverageScore <- cbind.data.frame(sequencesCombination, averageScore=meanScore)
114 ## SUPPRESSION METABOLITE AVEC SCORE MOYEN < SEUIL
115 ## sequencesCombinationAverageScoreSeuilFiltre <- filter(sequencesCombinationAverageScore, averageScore >= seuil)
116 sequencesCombinationAverageScoreSeuilFiltre <- sequencesCombinationAverageScore[sequencesCombinationAverageScore$averageScore > seuil, ]
117 }
118
119 return(list(COSY=annotationCOSY, HMBC=annotationHMBC, HSQC=annotationHSQC, JRES=annotationJRES, TOCSY=annotationTOCSY,
120 combination=sequencesCombinationAverageScoreSeuilFiltre))
121 }