Mercurial > repos > lain > xseekerpreparator

changeset 12:bdb2878ee189 draft

Find changesets by keywords (author, files, the commit message), revision number or hash, or revset expression.
" master branch Updating"
author lain
date Wed, 07 Apr 2021 13:05:36 +0000
parents f4fc4a0f41e2
children 26f01380145d
files XSeekerPreparator.R XSeekerPreparator.xml galaxy/tools/LC-MSMS/XSeekerPreparator.R galaxy/tools/LC-MSMS/XSeekerPreparator.xml
diffstat 4 files changed, 102 insertions(+), 30 deletions(-) [+]
line wrap: on
line diff
--- a/XSeekerPreparator.R	Thu Mar 18 10:46:01 2021 +0000
+++ b/XSeekerPreparator.R	Wed Apr 07 13:05:36 2021 +0000
@@ -484,18 +484,44 @@
     singlefile <- rdata$singlefile
     process_arg_list <- rdata$listOFlistArguments
     var_meta <- rdata$variableMetadata
-    ## We needed to get rid of the rdata, which is very big.
-    ## So we gathered all variable assignment from rdata here, and got rid of it.
-    rm(rdata)
 
     process_params <- list()
-    for (list_name in names(process_arg_list)) {
-        param_list <- list()
-        for (param_name in names(process_arg_list[[list_name]])) {
-            param_list[[param_name]] <- process_arg_list[[list_name]][[param_name]]
+    if (is.null(process_arg_list)) {
+        histories <- list()
+        for (history in xcms_set@.processHistory) {
+            if (
+                class(history@param) == "CentWaveParam"
+                && history@type == "Peak detection"
+            ) {
+                params <- history@param
+                process_params <- list(list(
+                    xfunction="annotatediff",
+                    ppm=params@ppm,
+                    peakwidth=sprintf("%s - %s", params@peakwidth[[1]], params@peakwidth[[2]]),
+                    snthresh=params@snthresh,
+                    prefilterStep=params@prefilter[[1]],
+                    prefilterLevel=params@prefilter[[2]],
+                    mzdiff=params@mzdiff,
+                    fitgauss=params@fitgauss,
+                    noise=params@noise,
+                    mzCenterFun=params@mzCenterFun,
+                    integrate=params@integrate,
+                    firstBaselineCheck=params@firstBaselineCheck,
+                    snthreshIsoROIs=!identical(params@roiScales, numeric(0))
+                ))
+                break
+            }
         }
-        process_params[[length(process_params)+1]] <- param_list
+    } else {
+        for (list_name in names(process_arg_list)) {
+            param_list <- list()
+            for (param_name in names(process_arg_list[[list_name]])) {
+                param_list[[param_name]] <- process_arg_list[[list_name]][[param_name]]
+            }
+            process_params[[length(process_params)+1]] <- param_list
+        }
     }
+
     message("Parameters from previous processes extracted.")
 
 
@@ -532,7 +558,7 @@
         env$tic <- ms_file@tic
         env$mz <- ms_file@env$mz
         env$scanindex <- ms_file@scanindex
-        env$scantime <- ms_file@scantime
+        env$scantime <- ms_file@scantime * 60
         env$intensity <- ms_file@env$intensity
         env$polarity <- as.character(ms_file@polarity[[1]])
 
@@ -555,6 +581,7 @@
     context$show_percent <- show_percent
     context$cluster_mean_rt_abundance <- list()
     context$central_feature <- list()
+    context$adducts <- list()
     load_variable_metadata(orm, var_meta, context)
     clusters <- context$clusters
     rm(context)
@@ -695,6 +722,7 @@
             }
         }
         create_associated_cluster(
+            orm,
             context$central_feature[[clusterID]],
             dummy_feature, clusterID,
             context, curent_var_meta, next_pc_group,
@@ -761,22 +789,31 @@
 }
 
 create_associated_cluster <- function(
+    orm,
     sample_no, feature, clusterID,
     context, curent_var_meta, next_pc_group, next_align_group
 ) {
-    pcgroup <- as.numeric(curent_var_meta[["pcgroup"]])
-    adduct <- as.character(curent_var_meta[["adduct"]])
-    annotation <- curent_var_meta[["isotopes"]]
     clusterID <- as.character(clusterID)
     if (is.null(cluster <- context$clusters[[clusterID]])) {
+        pcgroup <- as.numeric(curent_var_meta[["pcgroup"]])
+        adduct_name <- as.character(curent_var_meta[["adduct"]])
+        annotation <- curent_var_meta[["isotopes"]]
         cluster <- context$clusters[[clusterID]] <- orm$cluster(
             pc_group=pcgroup + next_pc_group,
-            adduct=adduct,
+            # adduct=adduct,
             align_group=next_align_group,
             # curent_group=curent_group,
             clusterID=context$clusterID,
             annotation=annotation
         )
+        if (is.null(adduct <- context$adducts[[adduct_name]])) {
+            context$adducts[[adduct_name]] <- orm$adduct()$load_by(name=adduct_name)$first()
+            if (is.null(adduct <- context$adducts[[adduct_name]])) {
+                adduct <- context$adducts[[adduct_name]] <- orm$adduct(name=adduct_name, charge=0)
+                adduct$save()
+            }
+        }
+        cluster$set_adduct(adduct)
         ## Crappy hack to assign sample id to cluster without loading the sample.
         ## Samples are too big (their sample$env) and slows the process, and eat all the menory
         ## so we dont't want to load them.
--- a/XSeekerPreparator.xml	Thu Mar 18 10:46:01 2021 +0000
+++ b/XSeekerPreparator.xml	Wed Apr 07 13:05:36 2021 +0000
@@ -197,13 +197,12 @@
             score="FLOAT",
             deviation="FLOAT",
             status="TEXT",
-            adduct="TEXT",
             curent_group="INTEGER",
             pc_group="INTEGER",
             align_group="INTEGER",
             xcms_group="INTEGER"
         ),
-        one=list("sample", "compound")
+        one=list("sample", "compound", "adduct")
     ),
     compound=DBModelR::ModelDefinition(
         table="compound",
--- a/galaxy/tools/LC-MSMS/XSeekerPreparator.R	Thu Mar 18 10:46:01 2021 +0000
+++ b/galaxy/tools/LC-MSMS/XSeekerPreparator.R	Wed Apr 07 13:05:36 2021 +0000
@@ -484,18 +484,44 @@
     singlefile <- rdata$singlefile
     process_arg_list <- rdata$listOFlistArguments
     var_meta <- rdata$variableMetadata
-    ## We needed to get rid of the rdata, which is very big.
-    ## So we gathered all variable assignment from rdata here, and got rid of it.
-    rm(rdata)
 
     process_params <- list()
-    for (list_name in names(process_arg_list)) {
-        param_list <- list()
-        for (param_name in names(process_arg_list[[list_name]])) {
-            param_list[[param_name]] <- process_arg_list[[list_name]][[param_name]]
+    if (is.null(process_arg_list)) {
+        histories <- list()
+        for (history in xcms_set@.processHistory) {
+            if (
+                class(history@param) == "CentWaveParam"
+                && history@type == "Peak detection"
+            ) {
+                params <- history@param
+                process_params <- list(list(
+                    xfunction="annotatediff",
+                    ppm=params@ppm,
+                    peakwidth=sprintf("%s - %s", params@peakwidth[[1]], params@peakwidth[[2]]),
+                    snthresh=params@snthresh,
+                    prefilterStep=params@prefilter[[1]],
+                    prefilterLevel=params@prefilter[[2]],
+                    mzdiff=params@mzdiff,
+                    fitgauss=params@fitgauss,
+                    noise=params@noise,
+                    mzCenterFun=params@mzCenterFun,
+                    integrate=params@integrate,
+                    firstBaselineCheck=params@firstBaselineCheck,
+                    snthreshIsoROIs=!identical(params@roiScales, numeric(0))
+                ))
+                break
+            }
         }
-        process_params[[length(process_params)+1]] <- param_list
+    } else {
+        for (list_name in names(process_arg_list)) {
+            param_list <- list()
+            for (param_name in names(process_arg_list[[list_name]])) {
+                param_list[[param_name]] <- process_arg_list[[list_name]][[param_name]]
+            }
+            process_params[[length(process_params)+1]] <- param_list
+        }
     }
+
     message("Parameters from previous processes extracted.")
 
 
@@ -532,7 +558,7 @@
         env$tic <- ms_file@tic
         env$mz <- ms_file@env$mz
         env$scanindex <- ms_file@scanindex
-        env$scantime <- ms_file@scantime
+        env$scantime <- ms_file@scantime * 60
         env$intensity <- ms_file@env$intensity
         env$polarity <- as.character(ms_file@polarity[[1]])
 
@@ -555,6 +581,7 @@
     context$show_percent <- show_percent
     context$cluster_mean_rt_abundance <- list()
     context$central_feature <- list()
+    context$adducts <- list()
     load_variable_metadata(orm, var_meta, context)
     clusters <- context$clusters
     rm(context)
@@ -695,6 +722,7 @@
             }
         }
         create_associated_cluster(
+            orm,
             context$central_feature[[clusterID]],
             dummy_feature, clusterID,
             context, curent_var_meta, next_pc_group,
@@ -761,22 +789,31 @@
 }
 
 create_associated_cluster <- function(
+    orm,
     sample_no, feature, clusterID,
     context, curent_var_meta, next_pc_group, next_align_group
 ) {
-    pcgroup <- as.numeric(curent_var_meta[["pcgroup"]])
-    adduct <- as.character(curent_var_meta[["adduct"]])
-    annotation <- curent_var_meta[["isotopes"]]
     clusterID <- as.character(clusterID)
     if (is.null(cluster <- context$clusters[[clusterID]])) {
+        pcgroup <- as.numeric(curent_var_meta[["pcgroup"]])
+        adduct_name <- as.character(curent_var_meta[["adduct"]])
+        annotation <- curent_var_meta[["isotopes"]]
         cluster <- context$clusters[[clusterID]] <- orm$cluster(
             pc_group=pcgroup + next_pc_group,
-            adduct=adduct,
+            # adduct=adduct,
             align_group=next_align_group,
             # curent_group=curent_group,
             clusterID=context$clusterID,
             annotation=annotation
         )
+        if (is.null(adduct <- context$adducts[[adduct_name]])) {
+            context$adducts[[adduct_name]] <- orm$adduct()$load_by(name=adduct_name)$first()
+            if (is.null(adduct <- context$adducts[[adduct_name]])) {
+                adduct <- context$adducts[[adduct_name]] <- orm$adduct(name=adduct_name, charge=0)
+                adduct$save()
+            }
+        }
+        cluster$set_adduct(adduct)
         ## Crappy hack to assign sample id to cluster without loading the sample.
         ## Samples are too big (their sample$env) and slows the process, and eat all the menory
         ## so we dont't want to load them.
--- a/galaxy/tools/LC-MSMS/XSeekerPreparator.xml	Thu Mar 18 10:46:01 2021 +0000
+++ b/galaxy/tools/LC-MSMS/XSeekerPreparator.xml	Wed Apr 07 13:05:36 2021 +0000
@@ -197,13 +197,12 @@
             score="FLOAT",
             deviation="FLOAT",
             status="TEXT",
-            adduct="TEXT",
             curent_group="INTEGER",
             pc_group="INTEGER",
             align_group="INTEGER",
             xcms_group="INTEGER"
         ),
-        one=list("sample", "compound")
+        one=list("sample", "compound", "adduct")
     ),
     compound=DBModelR::ModelDefinition(
         table="compound",