ezrun,uzh

plot dendrogram with colour in countQC

Better visualisation of groups in the dendrogram.

SCCOuntQC: unique gene symbols

library(scater)
rownames(sce) <- uniquifyFeatureNames(rowData(sce)$ENSEMBL, rowData(sce)$SYMBOL)

Replace samtools with "Tools/sambamba"

When parallelisation is possbile, "Tools/sambamba" has better performance.

introduce sgSeq Apps for alternative splicing

Add quasi-likelihood (QL) F-test option for glm

QL as the default.

Fisher's exact test for DEG

For the up- and downregulated genes separately, run a Fisher's exact
test to see if the regulated genes overlap with the high/low GC, short
long genes. i.e.

gcTypes = data.frame("GC < 0.4"=as.numeric(rowData(rawData)$gc) < 0.4,
                         "GC > 0.6"=as.numeric(rowData(rawData)$gc) > 0.6,
                         check.names=FALSE)
widthTypes = data.frame("width <
800nt"=as.numeric(rowData(rawData)$width) < 800,
                            "width >
2000nt"=as.numeric(rowData(rawData)$width) > 2000,
                            check.names=FALSE)

That should be part of the report. If there is a significant overlap.
The result is suspicious and potentially an artefact.

Unable to build Vignette

Hi,

What is the best way to build vignette for ezRun.

Different colour for NA in heatmaps of plate

fix emapplot from new clusterProfiler

emapplot from clusterProfiler is problematic.
Disable it for now.
It uses ggplot style igraph (ggraph) to plot the network. Need more time to debug

Example:
https://fgcz-sushi.uzh.ch/projects/p2729/DESeq2_26428_7--over--5_2018-05-06--12-42-45/scripts

10x like Median Genes per Cell saturant plot in SCCountQC

https://fgcz-gstore.uzh.ch/projects/p2220/CellRangerCount_35138_NOV86_o5444_2019-04-26--11-17-58/Differentiated_lung_cells/outs/web_summary.html

More space for multimatch plotly plot

RNABamStats

Fix the too wide table in RnaBamStats report

Match Count Percentage table.
Example:
https://fgcz-gstore.uzh.ch/projects/p2578/RNABamStats_22125_2017-11-06--13-42-26/RNA_BAM_Statistics/00index.html

ncPRO app requires "Read Count" column

When the "Read Count" column is absent from the dataset definition, the app stops with
Error in `$<-.data.frame`(`*tmp*`, removed, value = integer(0))

I suspect this is due too this part of the app:

ezRun/R/app-ncpro.R

Lines 88 to 100 in 6992d62

    
             if (!is.null(dataset$"Read Count") && is.numeric(dataset$"Read Count") && all(dataset$"Read Count" > 0)){ 
        
               readCounts$untrimmed = dataset[samples, "Read Count"]     
        
             } 
        
             readCounts$remaining = countReadsInFastq(trimmedFastqFiles) 
        
             ezWrite.table(readCounts, "trimCounts.txt") 
        
             readCounts$removed = readCounts$untrimmed - readCounts$remaining 
        
           #   plotCmd = expression({ 
        
           #     par(mar=c(12, 4.1, 4.1, 2.1))   
        
           #     barplot(t(as.matrix(readCounts[ , c("remaining", "removed")])), las=2, border=NA, 
        
           #             main="Read Counts after trimming", legend.text=TRUE, col=c("gray30", "gray")) 
        
           #   }) 
        
           #   unusedLink = ezImageFileLink(plotCmd, file=param$readCountsBarplot, width=400 + nrow(readCounts) * 10, height=700)

(More specifically line 93)

I haven't tried, but putting it under the line 88 conditional could probably solve the issue; alternatively, making Read Count a necessary column (e.g. in the sushi app) obviously does solve the problem.

aggregateCountsByGeneSE: don't do the conversion to array after aggregation.

create `transcripts.only.gtf` in the reference build pipeline

make FastQScreen work for Single Cell data

nReads may not be used in RnaBamStats

clean if not needed.

possible error in 'DADA2CreateSeqTab'

During installation,

Note: possible error in 'DADA2CreateSeqTab(sampleName, ': unused arguments (maxLen, minLen, isPaired)

setUnlistDataNames warnings

Why there are warnings from RnaBamStats app? possibly from getRangesCoverage.
The results looks fine.

40: In setUnlistDataNames(x@unlistData, x@partitioning, use.names,  ... :
  failed to set names on the unlisted CompressedRleList object

https://gatkforums.broadinstitute.org/gatk/discussion/11588/cigar-after-picard-mergebamalignment#latest

Differential expression test:	QL

Slingshot
Monocle
Identifying temporally expressed genes

setwd("/scratch/gtan/debug/p2710-CountQC")
library(ezRun)
param = list()
param[['cores']] = '1'
param[['ram']] = '2'
param[['scratch']] = '10'
param[['node']] = ''
param[['process_mode']] = 'DATASET'
param[['name']] = 'Count_QC'
param[['refBuild']] = 'p2710_Perma/FGCZ/Metabat_20180427/Annotation/Release_2018_04_30'
param[['refFeatureFile']] = 'genes.gtf'
param[['featureLevel']] = 'gene'
param[['normMethod']] = 'logMean'
param[['runGO']] = 'true'
param[['backgroundExpression']] = '10'
param[['transcriptTypes']] = ''
param[['specialOptions']] = ''
param[['expressionName']] = ''
param[['mail']] = '[email protected]'
param[['dataRoot']] = '/srv/gstore/projects'
param[['resultDir']] = 'p2710/CountQC_26335_2018-05-05--15-31-20'
output = list()
output[['Name']] = 'Count_QC'
output[['Species']] = 'NA'
output[['refBuild']] = 'p2710_Perma/FGCZ/Metabat_20180427/Annotation/Release_2018_04_30'
output[['Static Report [Link]']] = 'p2710/CountQC_26335_2018-05-05--15-31-20/Count_QC/00index.html'
output[['Live Report [Link]']] = 'http://fgcz-shiny.uzh.ch/fgcz_exploreCountQC_app/?data=p2710/CountQC_26335_2018-05-05--15-31-20/Count_QC/counts-qjliarzqgzfz-EzResult.RData'
output[['Report [File]']] = 'p2710/CountQC_26335_2018-05-05--15-31-20/Count_QC'
input = '/srv/gstore/projects/p2710/CountQC_26335_2018-05-05--15-31-20/input_dataset.tsv'
EzAppCountQC$new()$run(input=input, output=output, param=param)

	if (!is.null(dataset$"Read Count") && is.numeric(dataset$"Read Count") && all(dataset$"Read Count" > 0)){
	readCounts$untrimmed = dataset[samples, "Read Count"]
	}
	readCounts$remaining = countReadsInFastq(trimmedFastqFiles)
	ezWrite.table(readCounts, "trimCounts.txt")
	readCounts$removed = readCounts$untrimmed - readCounts$remaining

	# plotCmd = expression({
	# par(mar=c(12, 4.1, 4.1, 2.1))
	# barplot(t(as.matrix(readCounts[ , c("remaining", "removed")])), las=2, border=NA,
	# main="Read Counts after trimming", legend.text=TRUE, col=c("gray30", "gray"))
	# })
	# unusedLink = ezImageFileLink(plotCmd, file=param$readCountsBarplot, width=400 + nrow(readCounts) * 10, height=700)

uzh / ezrun Goto Github PK

ezrun's People

Contributors

Stargazers

Watchers

Forkers

ezrun's Issues

Recommend Projects

Recommend Topics

Recommend Org