Mercurial > repos > recetox > ramclustr

comparison ramclustr_wrapper.R @ 11:da7722f665f4 draft

Find changesets by keywords (author, files, the commit message), revision number or hash, or revset expression.
planemo upload for repository https://github.com/RECETOX/galaxytools/tree/master/tools/ramclustr commit bc3445f7c41271b0062c7674108f57708d08dd28
author recetox
date Thu, 30 May 2024 14:52:11 +0000
parents 2d94da58904b
children
comparison
equal deleted inserted replaced
10:2d94da58904b 11:da7722f665f4
1 store_output <- function(ramclustr_obj, 1 store_output <- function(ramclustr_obj,
2 output_merge_msp, 2 output_merge_msp,
3 output_spec_abundance, 3 output_spec_abundance,
4 msp_file) { 4 msp_file) {
5 RAMClustR::write.msp(ramclustr_obj, one.file = output_merge_msp) 5 RAMClustR::write.msp(ramclustr_obj, one.file = output_merge_msp)
6 write.table(ramclustr_obj$SpecAbund, 6 write.table(ramclustr_obj$SpecAbund,
7 file = output_spec_abundance, 7 file = output_spec_abundance,
8 row.names = TRUE, quote = FALSE, col.names = NA, sep = "\t" 8 row.names = TRUE, quote = FALSE, col.names = NA, sep = "\t"
9 ) 9 )
10 10
11 if (!is.null(msp_file)) { 11 if (!is.null(msp_file)) {
12 exp_name <- ramclustr_obj$ExpDes[[1]][which( 12 exp_name <- ramclustr_obj$ExpDes[[1]][which(
13 row.names(ramclustr_obj$ExpDes[[1]]) == "Experiment" 13 row.names(ramclustr_obj$ExpDes[[1]]) == "Experiment"
14 ), 1] 14 ), 1]
15 filename <- paste("spectra/", exp_name, ".msp", sep = "") 15 filename <- paste("spectra/", exp_name, ".msp", sep = "")
16 file.copy(from = filename, to = msp_file, overwrite = TRUE) 16 file.copy(from = filename, to = msp_file, overwrite = TRUE)
17 } 17 }
18 } 18 }
19 19
20 load_experiment_definition <- function(filename) { 20 load_experiment_definition <- function(filename) {
21 experiment <- RAMClustR::defineExperiment(csv = filename) 21 experiment <- RAMClustR::defineExperiment(csv = filename)
22 return(experiment) 22 return(experiment)
23 } 23 }
24 24
25 read_metadata <- function(filename) { 25 read_metadata <- function(filename) {
26 data <- read.csv(filename, header = TRUE, stringsAsFactors = FALSE) 26 data <- read.csv(filename, header = TRUE, stringsAsFactors = FALSE)
27 27
28 if (!"qc" %in% colnames(data)) { 28 if (!"qc" %in% colnames(data)) {
29 if ("sampleType" %in% colnames(data)) { 29 if ("sampleType" %in% colnames(data)) {
30 data$qc <- ifelse(data$sampleType == "qc", TRUE, FALSE) 30 data$qc <- ifelse(data$sampleType == "qc", TRUE, FALSE)
31 }
31 } 32 }
32 }
33 33
34 if (!"order" %in% colnames(data)) { 34 if (!"order" %in% colnames(data)) {
35 if ("injectionOrder" %in% colnames(data)) { 35 if ("injectionOrder" %in% colnames(data)) {
36 names(data)[names(data) == "injectionOrder"] <- "order" 36 names(data)[names(data) == "injectionOrder"] <- "order"
37 }
37 } 38 }
38 }
39 39
40 return(data) 40 return(data)
41 } 41 }
42 42
43 read_ramclustr_aplcms <- function(ms1_featuredefinitions = NULL, 43 read_ramclustr_aplcms <- function(ms1_featuredefinitions = NULL,
44 ms1_featurevalues = NULL, 44 ms1_featurevalues = NULL,
45 df_phenodata = NULL, 45 df_phenodata = NULL,
47 exp_des = NULL, 47 exp_des = NULL,
48 st = NULL, 48 st = NULL,
49 ensure_no_na = TRUE, 49 ensure_no_na = TRUE,
50 ms1_featuredefinitions_ext = "parquet", 50 ms1_featuredefinitions_ext = "parquet",
51 ms1_featurevalues_ext = "parquet") { 51 ms1_featurevalues_ext = "parquet") {
52 if (ms1_featuredefinitions_ext == "parquet") { 52 if (ms1_featuredefinitions_ext == "parquet") {
53 ms1_featuredefinitions <- arrow::read_parquet(ms1_featuredefinitions) 53 ms1_featuredefinitions <- arrow::read_parquet(ms1_featuredefinitions)
54 } else { 54 } else {
55 ms1_featuredefinitions <- read.csv(ms1_featuredefinitions, 55 ms1_featuredefinitions <- read.csv(ms1_featuredefinitions,
56 header = TRUE, sep = "\t" 56 header = TRUE, sep = "\t", check.names = FALSE
57 )
58 }
59
60 if (ms1_featurevalues_ext == "parquet") {
61 ms1_featurevalues <- arrow::read_parquet(ms1_featurevalues)
62 } else {
63 ms1_featurevalues <- read.csv(ms1_featurevalues,
64 header = TRUE,
65 sep = "\t", check.names = FALSE
66 )
67 }
68
69 if (!is.null(df_phenodata)) {
70 if (phenodata_ext == "csv") {
71 df_phenodata <- read.csv(
72 file = df_phenodata,
73 header = TRUE, check.names = FALSE
74 )
75 } else {
76 df_phenodata <- read.csv(
77 file = df_phenodata,
78 header = TRUE, check.names = FALSE, sep = "\t"
79 )
80 }
81 }
82 if (!is.null(exp_des)) {
83 exp_des <- load_experiment_definition(exp_des)
84 }
85
86 feature_values <- ms1_featurevalues[-1]
87 feature_values <- t(feature_values)
88 colnames(feature_values) <- ms1_featurevalues[[1]]
89
90 feature_definitions <- data.frame(ms1_featuredefinitions)
91
92 ramclustr_obj <- RAMClustR::rc.get.df.data(
93 ms1_featureDefinitions = feature_definitions,
94 ms1_featureValues = feature_values,
95 phenoData = df_phenodata,
96 ExpDes = exp_des,
97 st = st,
98 ensure.no.na = ensure_no_na
57 ) 99 )
58 } 100 return(ramclustr_obj)
59
60 if (ms1_featurevalues_ext == "parquet") {
61 ms1_featurevalues <- arrow::read_parquet(ms1_featurevalues)
62 } else {
63 ms1_featurevalues <- read.csv(ms1_featurevalues, header = TRUE, sep = "\t")
64 }
65
66 if (!is.null(df_phenodata)) {
67 if (phenodata_ext == "csv") {
68 df_phenodata <- read.csv(
69 file = df_phenodata,
70 header = TRUE, check.names = FALSE
71 )
72 } else {
73 df_phenodata <- read.csv(
74 file = df_phenodata,
75 header = TRUE, check.names = FALSE, sep = "\t"
76 )
77 }
78 }
79 if (!is.null(exp_des)) {
80 exp_des <- load_experiment_definition(exp_des)
81 }
82
83 feature_values <- ms1_featurevalues[-1]
84 feature_values <- t(feature_values)
85 colnames(feature_values) <- ms1_featurevalues[[1]]
86
87 feature_definitions <- data.frame(ms1_featuredefinitions)
88
89 ramclustr_obj <- RAMClustR::rc.get.df.data(
90 ms1_featureDefinitions = feature_definitions,
91 ms1_featureValues = feature_values,
92 phenoData = df_phenodata,
93 ExpDes = exp_des,
94 st = st,
95 ensure.no.na = ensure_no_na
96 )
97 return(ramclustr_obj)
98 } 101 }
99 102
100 apply_normalisation <- function(ramclustr_obj = NULL, 103 apply_normalisation <- function(ramclustr_obj = NULL,
101 normalize_method, 104 normalize_method,
102 metadata_file = NULL, 105 metadata_file = NULL,
103 qc_inj_range, 106 qc_inj_range,
104 p_cut, 107 p_cut,
105 rsq_cut, 108 rsq_cut,
106 p_adjust) { 109 p_adjust) {
107 batch <- NULL 110 batch <- NULL
108 order <- NULL 111 order <- NULL
109 qc <- NULL 112 qc <- NULL
110 113
111 if (normalize_method == "TIC") { 114 if (normalize_method == "TIC") {
112 ramclustr_obj <- RAMClustR::rc.feature.normalize.tic( 115 ramclustr_obj <- RAMClustR::rc.feature.normalize.tic(
113 ramclustObj = 116 ramclustObj =
114 ramclustr_obj 117 ramclustr_obj
115 ) 118 )
116 } else if (normalize_method == "quantile") { 119 } else if (normalize_method == "quantile") {
117 ramclustr_obj <- RAMClustR::rc.feature.normalize.quantile(ramclustr_obj) 120 ramclustr_obj <- RAMClustR::rc.feature.normalize.quantile(ramclustr_obj)
118 } else if (normalize_method == "batch.qc") { 121 } else if (normalize_method == "batch.qc") {
119 if (!(is.null(metadata_file) || metadata_file == "None")) { 122 if (!(is.null(metadata_file) || metadata_file == "None")) {
120 metadata <- read_metadata(metadata_file) 123 metadata <- read_metadata(metadata_file)
121 batch <- metadata$batch 124 batch <- metadata$batch
122 order <- metadata$order 125 order <- metadata$order
123 qc <- metadata$qc 126 qc <- metadata$qc
127 }
128
129 ramclustr_obj <- RAMClustR::rc.feature.normalize.batch.qc(
130 order = order,
131 batch = batch,
132 qc = qc,
133 ramclustObj = ramclustr_obj,
134 qc.inj.range = qc_inj_range
135 )
136 } else {
137 if (!(is.null(metadata_file) || metadata_file == "None")) {
138 metadata <- read_metadata(metadata_file)
139 batch <- metadata$batch
140 order <- metadata$order
141 qc <- metadata$qc
142 }
143
144 ramclustr_obj <- RAMClustR::rc.feature.normalize.qc(
145 order = order,
146 batch = batch,
147 qc = qc,
148 ramclustObj = ramclustr_obj,
149 p.cut = p_cut,
150 rsq.cut = rsq_cut,
151 p.adjust = p_adjust
152 )
124 } 153 }
125 154 return(ramclustr_obj)
126 ramclustr_obj <- RAMClustR::rc.feature.normalize.batch.qc(
127 order = order,
128 batch = batch,
129 qc = qc,
130 ramclustObj = ramclustr_obj,
131 qc.inj.range = qc_inj_range
132 )
133 } else {
134 if (!(is.null(metadata_file) || metadata_file == "None")) {
135 metadata <- read_metadata(metadata_file)
136 batch <- metadata$batch
137 order <- metadata$order
138 qc <- metadata$qc
139 }
140
141 ramclustr_obj <- RAMClustR::rc.feature.normalize.qc(
142 order = order,
143 batch = batch,
144 qc = qc,
145 ramclustObj = ramclustr_obj,
146 p.cut = p_cut,
147 rsq.cut = rsq_cut,
148 p.adjust = p_adjust
149 )
150 }
151 return(ramclustr_obj)
152 } 155 }