Skip to content

Commit

Permalink
fixed tests and namespaces
Browse files Browse the repository at this point in the history
  • Loading branch information
@hechth
hechth committed Jan 21, 2025
1 parent 605a45c commit 5adfebd
Show file tree
Hide file tree
Showing 3 changed files with 105 additions and 117 deletions.
155 changes: 74 additions & 81 deletions tools/bioconductor-scp/bioconductor_scp.xml
View file
Original file line number Diff line number Diff line change
Expand Up @@ -14,26 +14,21 @@
<requirement type="package" version="1.34.0">bioconductor-scater</requirement>
<requirement type="package" version="1.16.0">bioconductor-qfeatures</requirement>
<requirement type="package" version="3.62.1">bioconductor-limma</requirement>
<requirement type="package" version="2.2.1">r-ggplot2</requirement>
<requirement type="package" version="3.5.1">r-ggplot2</requirement>
</requirements>
<required_files>
<include path="utils.r" />
</required_files>
<expand macro="creator" />
<command detect_errors="exit_code"><![CDATA[
echo ${run_script} &&
Rscript -e 'source("${__tool_directory__}/utils.r")' -e 'source("${run_script}")'
#if $data_export.export_R_script
&& cat ${run_script} >> $script
#end if
]]></command>
<configfiles>
<configfile name="run_script"><![CDATA[
library("scp")
library("ggplot2")
library("dplyr")
library("sva")
library("impute")
data_input <- read.delim("$input_data", sep="\t")
metadata <- read.delim("$input_annotations", sep="\t")
runCol <- colnames(data_input)[$runcol]
Expand All @@ -49,28 +44,28 @@
removeEmptyCols = $remove_empty_columns
)
number_of_assays <- length(scp)
scp <- zeroIsNA(scp, i = 1:number_of_assays)
scp <- QFeatures::zeroIsNA(scp, i = 1:number_of_assays)
#if $filtering_data.filter_reverse
scp <- filterFeatures(scp,
scp <- QFeatures::filterFeatures(scp,
~ Reverse != "+")
#end if
#if $filtering_data.filter_contaminants
scp <- filterFeatures(scp,
scp <- QFeatures::filterFeatures(scp,
~ Potential.contaminant != "+")
#end if
scp <- filterFeatures(scp,
scp <- QFeatures::filterFeatures(scp,
~ !is.na(PIF) & PIF > ${filtering_data.PIF_threshold})
keepAssay <- dims(scp)[1, ] > ${filtering_data.minimum_features}
keepAssay <- QFeatures::dims(scp)[1, ] > ${filtering_data.minimum_features}
scp <- scp[, , keepAssay]
number_of_assays <- length(scp)
single_cell_channels <- gsub(",", "|", "${filtering_data.single_cells}")
scp <- computeSCR(scp,
scp <- scp::computeSCR(scp,
i = 1:number_of_assays,
colvar = "SampleType",
carrierPattern = "Carrier",
Expand All @@ -79,57 +74,57 @@
rowDataName = "MeanSCR")
#if $generate_QC_plots
QC_plot_SCR <- rbindRowData(scp, i = 1:number_of_assays) |>
QC_plot_SCR <- QFeatures::rbindRowData(scp, i = 1:number_of_assays) |>
data.frame() |>
ggplot(aes(x = MeanSCR)) +
geom_histogram() +
geom_vline(xintercept = c(1/$count_cell_carrier, 0.1),
ggplot2::ggplot(ggplot2::aes(x = MeanSCR)) +
ggplot2::geom_histogram() +
ggplot2::geom_vline(xintercept = c(1/$count_cell_carrier, 0.1),
lty = c(2, 1)) +
scale_x_log10()
ggplot2::scale_x_log10()
ggplot2::ggsave(filename = file.path("plots", "QC_plot_SCR.pdf"), QC_plot_SCR)
QC_plot_SCR_col <- rbindRowData(scp, i = 1:number_of_assays) |>
QC_plot_SCR_col <- QFeatures::rbindRowData(scp, i = 1:number_of_assays) |>
data.frame() |>
ggplot(aes(x = MeanSCR, color = runCol)) +
geom_density() +
geom_vline(xintercept = 0.02, lty = 2) +
geom_vline(xintercept = 1, lty = 1)+
scale_x_log10()
ggplot2::ggplot(ggplot2::aes(x = MeanSCR, color = runCol)) +
ggplot2::geom_density() +
ggplot2::geom_vline(xintercept = 0.02, lty = 2) +
ggplot2::geom_vline(xintercept = 1, lty = 1)+
ggplot2::scale_x_log10()
ggplot2::ggsave(filename = file.path("plots", "QC_plot_SCR_col.pdf"), QC_plot_SCR_col)
#end if
scp <- filterFeatures(scp,
scp <- QFeatures::filterFeatures(scp,
~ !is.na(MeanSCR) &
MeanSCR < ${filtering_data.SCR_threshold})
#if $filtering_data.qvalue_level == "PSM"
scp <- pep2qvalue(scp,
scp <- scp::pep2qvalue(scp,
i = 1:number_of_assays,
PEP = "dart_PEP",
rowDataName = "qvalue")
scp <- filterFeatures(scp,
scp <- QFeatures::filterFeatures(scp,
~ qvalue < ${filtering_data.qvalue_threshold})
#else
scp <- pep2qvalue(scp,
scp <- scp::pep2qvalue(scp,
i = 1:number_of_assays,
PEP = "dart_PEP",
groupBy = "$filtering_data.qvalue_level",
rowDataName = "qvalue")
scp <- filterFeatures(scp,
scp <- QFeatures::filterFeatures(scp,
~ qvalue < ${filtering_data.qvalue_threshold})
#end if
#if $filtering_data.divide_reference
scp <- divideByReference(scp,
scp <- scp::divideByReference(scp,
i = 1:number_of_assays,
colvar = "SampleType",
samplePattern = ".",
refPattern = "Reference")
#end if
scp <- aggregateFeaturesOverAssays(
scp <- scp::aggregateFeaturesOverAssays(
scp,
i = 1:number_of_assays,
fcol = fcol_aggregation_pep,
Expand All @@ -138,36 +133,36 @@
na.rm = TRUE
)
scp <- joinAssays(scp,
scp <- QFeatures::joinAssays(scp,
i = grep("peptide", names(scp)),
name = "peptides")
keep_samples <- unlist(strsplit("${peptide_filtering.samples_to_keep}", split=","))
scp <- scp[,colData(scp)[["SampleType"]] %in% keep_samples, ]
scp <- scp[,SummarizedExperiment::colData(scp)[["SampleType"]] %in% keep_samples, ]
#if $peptide_filtering.filter_median_intensity.cut_median_intensity == "yes"
medians <- colMedians(assay(scp[["peptides"]]), na.rm = TRUE)
colData(scp)[["MedianRI"]] <- medians
medians <- colMedians(SummarizedExperiment::assay(scp[["peptides"]]), na.rm = TRUE)
SummarizedExperiment::colData(scp)[["MedianRI"]] <- medians
#if $generate_QC_plots
QC_medianRI <- colData(scp) |>
QC_medianRI <- SummarizedExperimentcolData(scp) |>
data.frame() |>
ggplot() +
aes(x = MedianRI,
ggplot2::ggplot() +
ggplot2::aes(x = MedianRI,
y = SampleType,
fill = SampleType) +
geom_boxplot() +
scale_x_log10()
ggplot2::geom_boxplot() +
ggplot2::scale_x_log10()
ggplot2::ggsave(filename = file.path("plots", "QC_medianRI.pdf"), QC_medianRI)
#end if
scp <- scp[, !is.na(colData(scp)[["MedianRI"]]) & colData(scp)[["MedianRI"]] < ${peptide_filtering.filter_median_intensity.median_intensity_threshold}, ]
scp <- scp[, !is.na(SummarizedExperiment::colData(scp)[["MedianRI"]]) & SummarizedExperiment::colData(scp)[["MedianRI"]] < ${peptide_filtering.filter_median_intensity.median_intensity_threshold}, ]
#end if
#if $peptide_filtering.filter_median_CV.cut_median_CV == "yes"
number_of_observations <- ${peptide_filtering.filter_median_CV.minimum_peptides_CV}
CV_threshold <- ${peptide_filtering.filter_median_CV.median_CV_threshold}
scp <- medianCVperCell(scp,
scp <- scp::medianCVperCell(scp,
i = 1:number_of_assays,
groupBy = "Leading.razor.protein",
nobs = number_of_observations,
Expand All @@ -176,21 +171,21 @@
colDataName = "MedianCV")
#if $generate_QC_plots
QC_medianCV <- getWithColData(scp, "peptides") |>
colData() |>
QC_medianCV <- MultiAssayExperiment::getWithColData(scp, "peptides") |>
SummarizedExperiment::colData() |>
data.frame() |>
ggplot(aes(x = MedianCV,
fill = SampleType)) +
geom_boxplot() +
geom_vline(xintercept = CV_threshold)
ggplot2::ggplot(ggplot2::aes(x = MedianCV,
fill = SampleType)) +
ggplot2::geom_boxplot() +
ggplot2::geom_vline(xintercept = CV_threshold)
ggplot2::ggsave(filename = file.path("plots", "QC_medianCV.pdf"), QC_medianCV)
#end if
scp <- scp[, !is.na(colData(scp)[["MedianCV"]]) & colData(scp)[["MedianCV"]] < CV_threshold, ]
scp <- scp[, !is.na(SummarizedExperiment::colData(scp)[["MedianCV"]]) & SummarizedExperiment::colData(scp)[["MedianCV"]] < CV_threshold, ]
#end if
#if $peptide_filtering.remove_blank
scp <- scp[, colData(scp)[["SampleType"]] != "Blank", ]
scp <- scp[, SummarizedExperiment::colData(scp)[["SampleType"]] != "Blank", ]
#end if
#if $peptide_processing.normalization_method.choose_normalization == "simple"
Expand All @@ -201,28 +196,26 @@
method = "${peptide_processing.normalization_method.normalize_simple_method}"
)
#else
norm_function_col <- match.fun("$peptide_processing.normalization_method.normalize_columns")
scp <- sweep(
scp <- QFeatures::sweep(
scp,
i = "peptides",
MARGIN = 2,
FUN = "/",
STATS = norm_function_col(assay(scp[["peptides"]]), na.rm = TRUE),
STATS = ${peptide_processing.normalization_method.normalize_columns}(SummarizedExperiment::assay(scp[["peptides"]]), na.rm = TRUE),
name = "peptides_norm_col"
)
norm_function_row <- match.fun("$peptide_processing.normalization_method.normalize_rows")
scp <- sweep(
scp <- QFeatures::sweep(
scp,
i = "peptides_norm_col",
MARGIN = 1,
FUN = "/",
STATS = norm_function_row(assay(scp[["peptides_norm_col"]]), na.rm = TRUE),
STATS = ${peptide_processing.normalization_method.normalize_rows}(SummarizedExperiment::assay(scp[["peptides_norm_col"]]), na.rm = TRUE),
name = "peptides_norm"
)
#end if
scp <- logTransform(
scp <- QFeatures::logTransform(
scp,
base = ${peptide_processing.base},
i = "peptides_norm",
Expand All @@ -239,10 +232,10 @@
#if $peptide_processing.remove_missing_peptides.remove_peptides == "yes"
pNA <- ${peptide_processing.remove_missing_peptides.pNA_peptides} / 100
scp <- filterNA(scp, i = "peptides_log", pNA = pNA)
scp <- QFeatures::filterNA(scp, i = "peptides_log", pNA = pNA)
#end if
scp <- aggregateFeaturesOverAssays(
scp <- scp::aggregateFeaturesOverAssays(
scp,
i = "peptides_log",
fcol = fcol_aggregation_prot,
Expand All @@ -259,23 +252,21 @@
method = "${protein_processing.normalization_method_protein.normalize_simple_method_prot}"
)
#else
norm_function_col <- match.fun("${protein_processing.normalization_method_protein.normalize_columns_prot}")
scp <- sweep(
scp <- QFeatures::sweep(
scp,
i = "proteins",
MARGIN = 2,
FUN = "/",
STATS = norm_function_col(assay(scp[["proteins"]]), na.rm = TRUE),
STATS = ${protein_processing.normalization_method_protein.normalize_columns_prot}(SummarizedExperiment::assay(scp[["proteins"]]), na.rm = TRUE),
name = "proteins_norm_col"
)
norm_function_row <- match.fun("${protein_processing.normalization_method_protein.normalize_rows_prot}")
scp <- sweep(
scp <- QFeatures::sweep(
scp,
i = "proteins_norm_col",
MARGIN = 1,
FUN = "/",
STATS = norm_function_row(assay(scp[["proteins_norm_col"]]), na.rm = TRUE),
STATS = ${protein_processing.normalization_method_protein.normalize_rows_prot}(SummarizedExperiment::assay(scp[["proteins_norm_col"]]), na.rm = TRUE),
name = "proteins_norm"
)
#end if
Expand All @@ -292,7 +283,7 @@
dev.off()
#end if
scp <- impute(
scp <- QFeatures::impute(
scp,
i = "proteins_norm",
name = "proteins_imptd",
Expand All @@ -306,40 +297,40 @@
batch_colname <- colnames(metadata)[${batch_correction.select_batch_correction.batch_col}]
#if $batch_correction.select_batch_correction.batch_correction_method == "combat"
sce <- getWithColData(scp, "proteins_imptd")
batch <- colData(scp)[[batch_colname]]
model <- model.matrix(~ SampleType, data = colData(sce))
sce <- MultiAssayExperiment::getWithColData(scp, "proteins_imptd")
batch <- SummarizedExperiment::colData(scp)[[batch_colname]]
model <- stats::model.matrix(~ SampleType, data = SummarizedExperiment::colData(sce))
assay(sce) <- ComBat(
dat = assay(sce),
SummarizedExperiment::assay(sce) <- sva::ComBat(
dat = SummarizedExperiment::assay(sce),
batch = batch,
mod = model
)
scp <- addAssay(
scp <- QFeatures::addAssay(
scp,
y = sce,
name = "proteins_batchC"
)
scp <- addAssayLinkOneToOne(
scp <- QFeatures::addAssayLinkOneToOne(
scp,
from = "proteins_imptd",
to = "proteins_batchC"
)
#else
sce <- getWithColData(scp, "proteins_imptd")
sce <- MultiAssayExperiment::getWithColData(scp, "proteins_imptd")
preserve_colname <- colnames(metadata)[${batch_correction.select_batch_correction.preserve_col}]
assay(sce) <- limma::removeBatchEffect(
assay(sce),
SummarizedExperiment::assay(sce) <- limma::removeBatchEffect(
SummarizedExperiment::assay(sce),
group = sce[[preserve_colname]],
batch = sce[[batch_colname]]
)
scp <- addAssay(scp,
scp <- QFeatures::addAssay(scp,
y = sce,
name = "proteins_batchC")
scp <- addAssayLinkOneToOne(scp,
scp <- QFeatures::addAssayLinkOneToOne(scp,
from = "proteins_imptd",
to = "proteins_batchC")
#end if
Expand Down Expand Up @@ -386,8 +377,8 @@
#end if
#end if
assay_df <- as.data.frame(assay(scp, "proteins_batchC"))
row_metadata <- as.data.frame(rowData(scp[["proteins_batchC"]]))
assay_df <- as.data.frame(SummarizedExperiment::assay(scp, "proteins_batchC"))
row_metadata <- as.data.frame(SummarizedExperiment::rowData(scp[["proteins_batchC"]]))
export_data <- cbind(row_metadata, as.data.frame(assay_df))
write.table(export_data, file = '$Processed_data', sep = "\t", quote = F)
Expand Down Expand Up @@ -480,6 +471,7 @@
<param name="column_aggregation_proteins" value="17"/>
<param name="batch_col" value="2"/>
<param name="export_tables" value="true"/>
<param name="pca_coloring" value="4"/>
<output_collection name="intermediate_outputs" type="list" count="6"/>
<output name="Processed_data">
<assert_contents>
Expand All @@ -500,6 +492,7 @@
<param name="export_R_script" value="TRUE"/>
<param name="column_aggregation_peptides" value="6"/>
<param name="column_aggregation_proteins" value="17"/>
<param name="pca_coloring" value="4"/>
<output name="Processed_data">
<assert_contents>
<has_size size="625000" delta="20"/>
Expand All @@ -509,7 +502,7 @@
</output>
<output name="script">
<assert_contents>
<has_n_lines n="285"/>
<has_n_lines n="271"/>
<has_text text='ggplot2::ggsave(filename = file.path("plots", "PCA.pdf"), pca)'/>
</assert_contents>
</output>
Expand Down
Loading

0 comments on commit 5adfebd

Please sign in to comment.