kfukuta/wf-transcriptomes-v202
1
0
Fork 0
Code Issues Pull Requests Actions Packages Projects Releases Wiki Activity

Merge branch 'gh_8' into 'dev'

Browse Source 
Github issues

See merge request epi2melabs/workflow-containers/wf-isoforms!51
This commit is contained in:
Neil Horner 2022-04-11 15:08:47 +00:00
commit ebf814b4e2
5 changed files with 84 additions and 46 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

14
CHANGELOG.md
View File

@ -4,20 +4,22 @@ All notable changes to this project will be documented in this file.
The format is based on [Keep a Changelog](https://keepachangelog.com/en/1.0.0/),
and this project adheres to [Semantic Versioning](https://semver.org/spec/v2.0.0.html).
# Unreleased
## Added
- --direct_rna option.
## [v0.1.2]
### Added
- direct_rna option
- Some extra error handling
- Minor report display improvements
## [v0.1.1]
## Fixed
### Fixed
- Incorrect numbers and of transcripts caused by merging gff files with same gene and transcript ids
- Error handling in de novo pipeline. Skip clusters in build_backbones that cause an isONclust2 error
- Several small fixes in report plotting
## [v0.1.0]
## Added
### Added
- Added the denovo pipeline
## Changed
### Changed
- Updates to the report plots
## [v0.0.1]

79
bin/report.py
View File

@ -7,12 +7,12 @@ import math
from pathlib import Path
from aplanat import bars, hist, lines
from aplanat.components import fastcat
from aplanat.components import simple as scomponents
from aplanat.components.fastcat import read_length_plot, read_quality_plot
from aplanat.report import WFReport
from aplanat.util import Colors
from bokeh.layouts import gridplot
from bokeh.models import ColumnDataSource, Panel, Tabs
from bokeh.models import ColumnDataSource, Legend, Panel, Tabs
from bokeh.models.widgets import DataTable, TableColumn
from bokeh.palettes import Category10_10
from bokeh.plotting import figure
@ -253,6 +253,8 @@ def grouped_bar(df, title="", tilted_xlabs=False):
dodge_increment = abs(dodge_range[0] - dodge_range[1]) / \
(len(df.columns) - 1)
legend_it = []
if tilted_xlabs:
p.xaxis.major_label_orientation = math.pi / 4
@ -270,15 +272,18 @@ def grouped_bar(df, title="", tilted_xlabs=False):
i += 1
width = df.size / 60
p.vbar(
v = p.vbar(
x=dodge('x_groups', current_dodge, range=p.x_range), top=col,
width=width, source=source, color=color, legend_label=col)
width=width, source=source, color=color)
current_dodge += dodge_increment
legend_it.append([col, [v]])
legend = Legend(items=legend_it)
p.add_layout(legend, 'right')
p.x_range.range_padding = 0.1
p.xgrid.grid_line_color = None
p.legend.location = "top_left"
p.legend.orientation = "horizontal"
p.legend.orientation = "vertical"
return p
@ -319,6 +324,7 @@ def gff_compare_plots(report, gffcompare_outdirs: Path, sample_ids):
''')
tabs = []
gff_fails = False
for id_, dir_ in zip(sample_ids, gffcompare_outdirs):
stats, _, miss, novel, total = \
parse_gffcmp_stats(dir_ / 'str_merged.stats')
@ -333,16 +339,17 @@ def gff_compare_plots(report, gffcompare_outdirs: Path, sample_ids):
child=gridplot(
[bar_totals, bar_performance, bar_missed, bar_novel],
ncols=2, width=350, height=260), title=id_))
cover_panel = Tabs(tabs=tabs)
section.markdown(gffcompare_md)
section.plot(cover_panel)
else:
gffcompare_md = ('''
__Warning__: Gffcompare summary cannot be shown.
This could be due to incompatible reference fasta and gff
files.''')
section.markdown(gffcompare_md)
gff_fails = True
if gff_fails:
gffcompare_md += ('''
__Warning__: Some gffcompare summary cannot be shown.
This could be due to incompatible reference fasta and gff
files.''')
cover_panel = Tabs(tabs=tabs)
section.markdown(gffcompare_md)
section.plot(cover_panel)
names = {
'=': 'ExactMatch:=',
@ -598,9 +605,9 @@ def transcript_table(report, df_tmaps, covr_threshold):
### Query transcript table
Low coverage transcripts are removed to speed up the table viewing. <br>
This can be set with the parameter `transcript_table_cov_thresh` in the
config.
''')
Coverage threshold can be set with the parameter
`transcript_table_cov_thresh`.
''')
df = df_tmaps.drop(
columns=[
@ -614,6 +621,12 @@ def transcript_table(report, df_tmaps, covr_threshold):
df.sort_values('cov', ascending=True, inplace=True)
counts = list(range(len(df)))
# Filter on coverage threshold
df = df[df['cov'] >= covr_threshold]
if len(df) < 200: # Min size of table should be 200. Don't filter
df = df.sort_values('cov', ascending=False).iloc[:, 0:200]
covr_threshold = 0
# Keep Isoforms with coverage > threshold
vline_x = np.argmax(df['cov'] > covr_threshold)
vline_y = [0, df['cov'].max()]
@ -629,10 +642,6 @@ def transcript_table(report, df_tmaps, covr_threshold):
colors=['blue', 'red'])
section.plot(cov_plt)
# Filter on coverage threshold
df = df[df['cov'] >= covr_threshold]
if len(df) < 200: # Min size of table should be 200
df = df.sort_values('cov', ascending=False).iloc[:, 0:200]
# Make a column of number of isoforms in parent gene
gb = df.groupby(['ref_gene_id', 'sample_id']).count()
@ -707,6 +716,8 @@ def transcriptome_summary(report, gffs, sample_ids, denovo=False):
bar_isos = hist.histogram(
[isoforms_per_gene], colors=[Colors.cerulean],
title="isoforms per gene")
bar_isos.xaxis.axis_label = "Num. isoforms"
bar_isos.yaxis.axis_label = "Num. genes"
bar_isos.xaxis.major_label_orientation = math.pi / 2.8
plots.append(bar_isos)
@ -723,6 +734,8 @@ def transcriptome_summary(report, gffs, sample_ids, denovo=False):
fig = figure(title="Exons per transcript")
fig.vbar(
x, top=list(y), color=Colors.cerulean)
fig.xaxis.axis_label = 'Num. exons'
fig.yaxis.axis_label = 'Num. genes'
fig.xaxis.major_label_orientation = math.pi / 2.8
plots.append(fig)
@ -765,6 +778,23 @@ def load_sample_data(files, sample_ids, read_func=None):
return df_
def seq_stats_tabs(report, sample_ids, stats):
"""Make tabs of sequence summaries by sample."""
tabs = []
for id_, summ in sorted(zip(sample_ids, stats)):
df_sum = pd.read_csv(summ, index_col=False, sep='\t')
rlp = read_length_plot(df_sum)
rqp = read_quality_plot(df_sum)
grid = gridplot(
[rlp, rqp], ncols=2, sizing_mode="stretch_width")
tabs.append(Panel(child=grid, title=id_))
section = report.add_section()
section.markdown("""
### Sequence summaries""")
section.plot(Tabs(tabs=tabs))
def main():
"""Run the entry point."""
parser = argparse.ArgumentParser()
@ -815,12 +845,7 @@ def main():
revision=args.revision, commit=args.commit)
# Add reads summary section
for id_, summ in zip(sample_ids, args.summaries):
report.add_section(
section=fastcat.full_report(
[summ],
header='#### Read stats: {}'.format(id_)
))
seq_stats_tabs(report, args.sample_ids, args.summaries)
if args.alignment_stats is not None:
df_aln_stats = load_sample_data(args.alignment_stats, sample_ids)

31
main.nf
View File

@ -130,21 +130,32 @@ process split_bam{
output:
tuple val(sample_id), path('*.bam'), emit: bundles
script:
if (params["bundle_min_reads"] != false)
"""
"""
n=`samtools view -c $bam`
if [[ n -lt 1 ]]
then
echo 'There are no reads mapping for $sample_id. Exiting!'
exit 1
fi
re='^[0-9]+\$'
if [[ $params.bundle_min_reads =~ \$re ]]
then
echo "Bundling up the bams"
seqkit bam -j ${params.threads} -N ${params.bundle_min_reads} ${bam} -o bam_bundles/
mv bam_bundles/* .
let i=1
for b in *.bam; do
newname="${sample_id}_batch_\${i}.bam"
mv \$b \$newname
((i++))
for b in bam_bundles/*.bam; do
echo \$b
newname="${sample_id}_batch_\${i}.bam"
mv \$b \$newname
((i++))
done
"""
else
"""
echo 'no bundling'
ln -s ${bam} ${sample_id}_batch_1.bam
"""
fi
"""
}

4
nextflow.config
View File

@ -23,7 +23,7 @@ params {
sample = null
sample_sheet = null
sanitize_fastq = false
wfversion = "v0.1.1"
wfversion = "v0.1.2"
aws_image_prefix = null
aws_queue = null
report_name = "report"
@ -34,7 +34,7 @@ params {
schema_ignore_params = 'show_hidden_params,validate_params,monochrome_logs,aws_queue,aws_image_prefix,wfversion'
// Process cDNA reads using pychopper, turn off for direct RNA:
direct_rna = true
direct_rna = false
// Options passed to pychopper:
pychopper_opts = "-m edlib"

2
nextflow_schema.json
View File

@ -280,7 +280,7 @@
},
"wfversion": {
"type": "string",
"default": "v0.1.1",
"default": "v0.1.2",
"hidden": true
},
"monochrome_logs": {