From c183f739b93dab0204cad4f88dbec26a68afa42c Mon Sep 17 00:00:00 2001
From: Holger Brandl <brandl@mpi-cbg.de>
Date: Mon, 9 Mar 2015 13:57:00 +0100
Subject: [PATCH] cont. chipseq qc
---
chipseq_workflow/chipseq_utils.sh | 19 +++++++++++++------
dge_workflow/dge_utils.sh | 3 +--
dge_workflow/fastqc_summary.R | 8 ++++----
3 files changed, 18 insertions(+), 12 deletions(-)
diff --git a/chipseq_workflow/chipseq_utils.sh b/chipseq_workflow/chipseq_utils.sh
index cbd6bc3..75ceae0 100755
--- a/chipseq_workflow/chipseq_utils.sh
+++ b/chipseq_workflow/chipseq_utils.sh
@@ -49,24 +49,31 @@ scale_fill_discrete <- function (...){ scale_color_brewer(..., type = "seq", pal
#+ fig.height=nrow(algnSummary)
ggplot(algnSummary, aes(condition, mapping_efficiency)) +
geom_bar(stat="identity") +
- coord_flip() +
ylim(0,100) +
+ coord_flip() +
ggtitle("mapping efficiency")
ggplot(algnSummary, aes(condition, num_reads)) +
geom_bar(stat="identity") +
coord_flip() +
- ggtitle("read counts") +scale_y_continuous(labels=comma)
+ ggtitle("read counts") +
+ scale_y_continuous(labels=comma)
ggplot(algnSummary, aes(condition, unique_mapper_prop)) +
- geom_bar(stat="identity") + coord_flip() +
+ geom_bar(stat="identity") +
+ coord_flip() +
ggtitle("unique-mapper proportions") +
- scale_y_continuous(labels=comma)
+ scale_y_continuous(labels=percent) +
+ ylim(0,100)
+
ggplot(algnSummary, aes(condition, multi_mappers_prop)) +
- geom_bar(stat="identity") + coord_flip() +
+ geom_bar(stat="identity") +
+ coord_flip() +
ggtitle("multi-mapper proportions") +
- scale_y_continuous(labels=comma)
+ scale_y_continuous(labels=percent) +
+ ylim(0,100)
+
#> ## Bam Correlation
diff --git a/dge_workflow/dge_utils.sh b/dge_workflow/dge_utils.sh
index d1b18b7..2e6b9c6 100755
--- a/dge_workflow/dge_utils.sh
+++ b/dge_workflow/dge_utils.sh
@@ -5,8 +5,7 @@
source <(curl https://dl.dropboxusercontent.com/u/113630701/datautils/bash/lsf_utils.sh 2>&1 2>/dev/null)
## enable snippet spinning
-source <(curl https://dl.dropboxusercontent.com/u/113630701/datautils/R/utils/spinr.sh 2>&1 2>/dev/null)
-
+source <(curl https://raw.githubusercontent.com/holgerbrandl/datautils/master/R/spinr/spin_utils.sh 2>&1 2>/dev/null)
## define common binaries
export PATH=/projects/bioinfo/holger/bin/bowtie2-2.2.2:$PATH
diff --git a/dge_workflow/fastqc_summary.R b/dge_workflow/fastqc_summary.R
index a402dc3..2061feb 100755
--- a/dge_workflow/fastqc_summary.R
+++ b/dge_workflow/fastqc_summary.R
@@ -52,7 +52,7 @@ readCount <- function(statsFile){
readCounts <- fastqDataFiles %>% ldply(readCount) # %>% print_head()
require.auto(scales)
-#+ fig.width=12, fig.height=round(nrow(readCounts)/3)
+#+ fig.width=12, fig.height=round(nrow(readCounts)/2)
ggplot(readCounts, aes(run, num_reads)) + geom_bar(stat="identity") + coord_flip() + scale_y_continuous(labels=comma) + ggtitle("read counts")
# #+ results='asis'
@@ -75,7 +75,7 @@ readSummary <- function(statsFile){
qcSummary <- fastqDataFiles %>% ldply(readSummary)
#' # Base Quality Distribution Summary
-#+ fig.height=2+round(nrow(readCounts)/3), fig.width=12
+#+ fig.height=2+round(nrow(readCounts)/2), fig.width=12
qcSummary %>% ggplot(aes(score, run, fill=tolower(flag))) +
geom_tile() +
rotXlab() +
@@ -94,7 +94,7 @@ dupLevels <- fastqDataFiles %>% ldply(function(statsFile){
})
require.auto(scales)
-#+ fig.width=12, fig.height=round(nrow(dupLevels)/3)
+#+ fig.width=12, fig.height=round(nrow(dupLevels)/2)
ggplot(dupLevels, aes(run, dedup_proportion)) + geom_bar(stat="identity") + coord_flip() + scale_y_continuous(labels=percent) + ggtitle("unique_reads/total_reads") + ylim(0,1)
@@ -133,7 +133,7 @@ baseQualities %>% ggplot(aes(reorder(Base, base_order), Mean, group=run, color=r
runs <- with(baseQualities, as.data.frame(table(run)))
-#+ warning=FALSE, fig.width=15, fig.height=3*ceiling(nrow(runs)/3)
+#+ warning=FALSE, fig.width=15, fig.height=3*ceiling(nrow(runs)/2)
## http://stackoverflow.com/questions/12518387/can-i-create-an-empty-ggplot2-plot-in-r
baseQualities %>%
--
GitLab