Pure python scripts to make sashimi plots
Why am I write this?
Cause I trying to integrate sashimi plots within Flask, rather than embed JBrowse into it.
Thanks to ggsashimi, I learned how to extract junctions and so on from BAM/SAM files
- sashmi plot share y axis
-
Sahimi do not share y axis
-
Line plot, used to viz ATAC
Software requirements
click==7.0 cycler==0.10.0 et-xmlfile==1.0.1 filetype==1.0.5 jdcal==1.4.1 kiwisolver==1.1.0 matplotlib==3.0.1 numpy==1.15.4 openpyxl==2.5.9 pyparsing==2.4.2 pysam==0.15.1 python-dateutil==2.8.0 six==1.12.0 tqdm==4.28.1
- from source
git clone https://github.com/ygidtu/pysashimi.git cd pysashimi pip install -r requirements.txt - using docker image
git clone https://github.com/ygidtu/pysashimi.git cd pysashimi docker build --rm -t chenlinlab/sashimi .
There are three different mode in this suite of scripts
- normal -> quite like ggsashimi
- pipeline -> specific format of meta info is required
- no-bam -> draw sashimi plot without BAM, use exons to replace BAM density
- line -> draw a line plot which a used to viz ATAC-seq
This suite of scripts will automatically check the index of BAM files and gtf files.
If there are lack of any index file (.bai or .tbi), this script will try to create one. Therefore, the write permission may required.
> python main.py # docker run -it chenlinlab/sashimi
Usage: main.py [OPTIONS] COMMAND [ARGS]...
Welcome
This function is used to test the function of sashimi plotting
Options:
--version Show the version and exit.
-h, --help Show this message and exit.
Commands:
no-bam This function is used to plot sashimi without BAM file
normal This function is used to plot single sashimi plotting
pipeline This function is used to plot sashimi based on specific meta...- -e/--event: The region you want to plot, support chr1:100-200:+ or chr1:100-200:+@chr1:100-300:+.
- -g/--gtf: for now only gtf is supported, gff3 is also in the plan (quite easy though). This suite of scripts will try to create .tbi index for gtf file to increase the I/O speed first time. Therefore write permission of the directory of gtf file may required.
- -o/--output:
- For normal and no-bam: path to the output file, support common formats like pdf, png, svg, jpg, tif (may face compression issues, due to linux system do not install libtiff normally) and so on.
- For pipelin: path to the output directory, all the figures will saved as pdf
- --indicator-lines:
- For normal and no-bam: add additional vertical dash lines in figure, to show the splice site or some else. eg: 150,170.
- For pipeline: this only is a flag, no comma separated integers required
- --share-y: see samples
- --dpi: the resolution of output figure
- -t/--threshold: threshold to filter low abundance junctions, means will not draw any low expression juncitons
- --no-gene: flag, default this plot will add gene id above transcript id. If this is set, will only draw transcript id
- --color-factor: three modes support different kind of meta info, use this parameter to specify which column in meta info is used to assign colors
>> python main.py normal -h
Usage: main.py normal [OPTIONS]
This function is used to plot single sashimi plotting
Options:
-e, --event TEXT Event range eg: chr1:100-200:+ [required]
-b, --bam PATH Path to input BAM file.
Or a tab separated
text file,
- first column is path to BAM
file,
- second column is BAM file
alias(optional)
-g, --gtf PATH Path to gtf file, both transcript and exon
tags are necessary
-o, --output PATH Path to output graph file
--config PATH Path to config file, contains graph settings
of sashimi plot [default: /Users/zhangyiming
/Code/pysashimi/settings.ini]
-t, --threshold INTEGER RANGE Threshold to filter low abundance junctions
[default: 0]
-d, --dpi INTEGER RANGE The resolution of output file [default: 300]
--indicator-lines TEXT Where to plot additional indicator lines,
comma separated int
--share-y Whether different sashimi plots shared same y
axis
--no-gene Do not show gene id next to transcript id
--color-factor INTEGER RANGE Index of column with color levels (1-based)
[default: 1]
--log [0|2|10|zscore] y axis log transformed, 0 -> not log
transform; 2 -> log2; 10 -> log10
--customized-junction TEXT Path to junction table column name needs to
be bam name or bam alias.
-p, --process INTEGER RANGE How many cpu to use
--sort-by-color Whether sort input bam order, for better
looking
--share-y-by INTEGER Index of column with share y axis (1-based),
Need --share-y\.
For example, first 3 bam
files use same y axis, and the rest use
another [default: -1]
-h, --help Show this message and exit.
-
-b/--bam: two different types of parameter were supported
- path to specific BAM file
- path to a list of BAM files (tab separated), and at least one column in this file
- path to BAM files
- The alias of BAM files, due to this list is tab separated, so space in this alias is also fine.
- additional colmuns, may used to assign colors and so on
for example (do not add any table header in the actual list):
first column second column third column /home/user/test.bam Sample 1 Red -
If bam list file is used, the order of bam file in this list is used. If the
--sort-by-colorused, then reorder the bam list by colors, to make sure same color bam file will be together. -
-p/--process: multiple process to read data from input files for faster processing -
--share-y-by: make different input files use different y axis. For example:
-
--log: zscore is usedscipy.stats.zscoreto convert density to zscore, just for test usage, please no use it in final plots.
>> python main.py pipeline -h
Usage: main.py pipeline [OPTIONS]
This function is used to plot sashimi based on specific meta info
required a specific format of input file
This function is used to test the function of sashimi plotting
Options:
-i, --input PATH Path to the meta info [xlsx] [required]
-s, --span TEXT To span the input region,
int -> span
corresponding bp
float -> span by
percentage of input region [default: 100]
-g, --gtf PATH Path to gtf file, both transcript and exon
tags are necessary
-o, --output PATH Path to output directory
--config PATH Path to config file, contains graph settings
of sashimi plot [default: /Users/zhangyiming
/Code/pysashimi/settings.ini]
-t, --threshold INTEGER RANGE Threshold to filter low abundance junctions
[default: 0]
-d, --dpi INTEGER RANGE The resolution of output file [default: 300]
--indicator-lines Where to plot additional indicator lines
--share-y Whether different sashimi plots shared same y
axis
--no-gene Do not show gene id next to transcript id
--color-factor INTEGER RANGE Index of column with color levels (1-based)
[default: 1]
--log [0|2|10|zscore] y axis log transformed, 0 -> not log
transform; 2 -> log2; 10 -> log10
--customized-junction TEXT Path to junction table column name needs to
be bam name or bam alias.
-p, --process INTEGER RANGE How many cpu to use
--sort-by-color Whether sort input bam order, for better
looking
-h, --help Show this message and exit.- -i/--input: path to a two sheet xlsx file, see docs/sample.xlsx, the value (if is not empty) in first sheet will added to the top-right of each sashimi
- -s/--span:
- Int: will extand the input region by bp before drawing
- Float: will extand the input region by percentage of exist region length before drawing
- --indicator-lines: will extract the splice sites from the input splice event ids
>> python main.py no-bam -h
Usage: main.py no-bam [OPTIONS]
This function is used to plot sashimi without BAM file
Options:
-e, --event TEXT Event range eg: chr1:100-200:+ [required]
-i, --input PATH Path to junctions count table [required]
--required PATH Path to tab separated list file
1. the
column use to plot sashimi, identical with
count table column names
2. optional, the
alias of 1st column
3. additional columns
-g, --gtf PATH Path to gtf file, both transcript and exon
tags are necessary
-o, --output PATH Path to output graph file
--config PATH Path to config file, contains graph settings
of sashimi plot [default: /Users/zhangyiming
/Code/pysashimi/settings.ini]
-t, --threshold INTEGER RANGE Threshold to filter low abundance junctions
[default: 0]
-d, --dpi INTEGER RANGE The resolution of output file [default: 300]
--indicator-lines TEXT Where to plot additional indicator lines,
comma separated int
--share-y Whether different sashimi plots shared same y
axis
--no-gene Do not show gene id next to transcript id
--color-factor INTEGER RANGE Index of column with color levels (1-based)
[default: 1]
-h, --help Show this message and exit.- -i/--input: path to extracted count table
- --required: path to a list, quite like bam list in normal mode
- First column is the sample names used for plotting, corresponding to the column names of count table
- Second column is the alias names
- Additional columns