miami_generator is a python tool to create beautiful Miami plots (which are a
combination of two Manhattan plots into a single image, showing the results
from two distinct group).
The tool requires Python version 3 and recent versions of numpy, matplotlib and pandas.
The best way to install the tool is in a virtual python environment. The
following commands should work, considering that $PATH_TO_VENV is the path to
the virtual environment root directory, and $PATH_TO_SETUP is the path to the
directory containing the setup.py file for the tool.
. "$PATH_TO_VENV"/bin/activate
pip install -e "$PATH_TO_SETUP"$ miami_generator --help
usage: miami_generator [-h] [--data FILE] [--up-data FILE]
[--up-data-label LABEL] [--down-data FILE]
[--down-data-label LABEL] [--separator SEP]
[--chromosome CHR] [--position POS] [--p-value P]
[--group COL] [--output FILE] [--format FORMAT]
[--dpi INT] [--up-colors COL COL]
[--down-colors COL COL] [--chromosome-box-color COLOR]
[--significant-color COLOR] [--graph-title TITLE]
[--axis-text-size INT] [--chr-text-size INT]
[--label-text-size INT] [--group-text-size INT]
[--graph-width WIDTH] [--graph-height HEIGHT]
[--abline P [P ...]] [--point-size SIZE]
[--significant-point-size SIZE]
[--chromosome-spacing VALUE]
[--significant-threshold FLOAT] [--max-p-value FLOAT]
[--p-value-sampling]
[--p-value-sampling-threshold FLOAT]
[--p-value-sampling-proportion FLOAT]
Creates a Miami plot from two datasets.
optional arguments:
-h, --help show this help message and exit
Input options:
--data FILE A file containing two datasets. You need to specify
the column which stratifies the two datasets.
--up-data FILE A file containing one of the two datasets. This
dataset will be at the top of the Miami plot.
--up-data-label LABEL
The label for the datasets that will be at the top of
the Miami plot. If this value is not set and the data
comes from a single file, the label is automatically
generated.
--down-data FILE A file containing one of the two datasets. This
dataset will be at the bottom of the Miami plot.
--down-data-label LABEL
The label for the datasets that will be at the bottom
of the Miami plot. If this value is not set and the
data comes from a single file, the label is
automatically generated.
Data format:
--separator SEP The file(s) field separator. [TAB]
--chromosome CHR The column containing the chromosomes. [chr]
--position POS The column containing the positions. [pos]
--p-value P The column containing the p-values. [p]
--group COL The column containing the groups (for single file
dataset).
Output options:
--output FILE The name of the output file. The extension will be
added if missing. [miami]
--format FORMAT The format of the plot (pdf, png). [png]
--dpi INT The quality of the output (in dpi). [600]
Graph color:
--up-colors COL COL The two colors (odd and even chromosomes) for the
upper part of the Miami plot.
--down-colors COL COL
The two colors (odd and even chromosomes) for the
upper part of the Miami plot.
--chromosome-box-color COLOR
The COLOR for the box surrounding even chromosome
numbers. [#E5E5E5]
--significant-color COLOR
The color for points representing significant points.
[#FF0000]
Graph text:
--graph-title TITLE The TITLE of the graph.
--axis-text-size INT The axis font size. [10]
--chr-text-size INT The axis font size. [10]
--label-text-size INT
The axis font size. [12]
--group-text-size INT
The group label font size. [8]
Graph look:
--graph-width WIDTH The width of the graph, in inches. [14]
--graph-height HEIGHT
The height of the graph, in inches. [4]
--abline P [P ...] The p-value(s) at which a horizontal line should be
drawn.
--point-size SIZE The size of each points. [0.5]
--significant-point-size SIZE
The size of each significant points. [4.0]
--chromosome-spacing VALUE
The spacing between two chromosomes. [2.5E+07]
--significant-threshold FLOAT
The significant threshold for linkage or association.
[5.0E-08]
Optimization:
--max-p-value FLOAT Prevent plotting p-values higher than this cutoff.
[1.0]
--p-value-sampling Sample a subset of points for specific p-values (see '
--p-value-sampling-threshold' and '--p-value-sampling-
proportion'). [False]
--p-value-sampling-threshold FLOAT
Points with p-values equal or higher than this
threshold will be sampled. [0.01]
--p-value-sampling-proportion FLOAT
The proportion of points with a p-values equal or
higher than the threshold which will be kept for
plotting. [0.1]As an example, we are using GWAS results which were generated for both men and women.
miami_generator \
--up-data ./data/males_results.tsv \
--up-data-label Males \
--down-data ./data/females_results.tsv \
--down-data-label Females \
--chromosome CHR \
--position BP \
--p-value "P(R)" \
--chromosome-spacing 25e6 \
--max-p-value 0.5 \
--output ./example.png \
--graph-title "Meta-analysis for myalgia and CK"The following plot was generated.
If the data was present in a single file (with a column describing the two groups), the following command would have generated the same image.
miami_generator \
--data ./data/results.tsv \
--chromosome CHR \
--position BP \
--group SEX \
--p-value "P(R)" \
--chromosome-spacing 25e6 \
--max-p-value 0.5 \
--output ./example.png \
--graph-title "Meta-analysis for myalgia and CK"