Scientific Data Visualization

Create publication-quality scientific plots and visualizations using matplotlib and seaborn.

Overview

This skill provides data visualization capabilities for scientific data:

• Line plots, scatter plots, bar charts
• Heatmaps and clustermaps
• Box plots and violin plots
• Histograms and density plots
• Sequence logos (for bioinformatics)
• Multiple subplot layouts

Usage

Create a line plot from CSV:

python3 {baseDir}/scripts/plot_data.py line --data data.csv --x time --y value --output plot.png

Create a scatter plot:

python3 {baseDir}/scripts/plot_data.py scatter --data data.csv --x x_col --y y_col --hue group

Create a heatmap:

python3 {baseDir}/scripts/plot_data.py heatmap --data matrix.csv --output heatmap.png

Create a bar chart:

python3 {baseDir}/scripts/plot_data.py bar --data data.csv --x category --y value

Plot from JSON data:

python3 {baseDir}/scripts/plot_data.py line --json '{"x": [1,2,3], "y": [4,5,6]}'

Plot Types

line

Line plot for continuous data.

Parameter	Description	Default
--data	CSV file path	-
--json	JSON data string	-
--x	X-axis column	Required
--y	Y-axis column(s), comma-separated	Required
--hue	Color grouping column	-
--style	Line style column	-
--markers	Add markers	False

scatter

Scatter plot for showing relationships.

Parameter	Description	Default
--data	CSV file path	-
--x	X-axis column	Required
--y	Y-axis column	Required
--hue	Color grouping column	-
--size	Size column	-
--alpha	Point transparency	0.7

bar

Bar chart for categorical data.

Parameter	Description	Default
--data	CSV file path	-
--x	Category column	Required
--y	Value column	Required
--hue	Color grouping column	-
--horizontal	Horizontal bars	False
--error	Error bar column	-

heatmap

Heatmap for matrix data.

Parameter	Description	Default
--data	CSV file path	Required
--cmap	Color map	viridis
--annotate	Show values	False
--cluster	Cluster rows/columns	False

box

Box plot for distributions.

Parameter	Description	Default
--data	CSV file path	-
--x	Grouping column	-
--y	Value column	Required
--hue	Color grouping column	-

violin

Violin plot for distributions.

Parameter	Description	Default
--data	CSV file path	-
--x	Grouping column	-
--y	Value column	Required
--hue	Color grouping column	-
--split	Split violins by hue	False

histogram

Histogram for distributions.

Parameter	Description	Default
--data	CSV file path	-
--x	Value column	Required
--bins	Number of bins	auto
--kde	Add KDE line	False
--hue	Color grouping column	-

Common Options

Option	Description	Default
--output	Output file path	plot.png
--format	Output format: png, svg, pdf	png
--title	Plot title	-
--xlabel	X-axis label	column name
--ylabel	Y-axis label	column name
--figsize	Figure size (width,height)	10,6
--style	Seaborn style	whitegrid
--palette	Color palette	deep
--dpi	Output resolution	150
--legend	Legend position	auto
--logx	Log scale X-axis	False
--logy	Log scale Y-axis	False

Examples

Multi-line plot with legend:

python3 {baseDir}/scripts/plot_data.py line --data timeseries.csv --x date --y "temp,humidity" --title "Weather Data" --output weather.png

Scatter plot with regression line:

python3 {baseDir}/scripts/plot_data.py scatter --data experiment.csv --x dose --y response --hue treatment --title "Dose Response" --output dose_response.png

Clustered heatmap:

python3 {baseDir}/scripts/plot_data.py heatmap --data expression.csv --cluster --cmap RdBu_r --title "Gene Expression" --output heatmap.svg --format svg

Box plot with multiple groups:

python3 {baseDir}/scripts/plot_data.py box --data measurements.csv --x condition --y value --hue treatment --title "Treatment Effects"

Histogram with KDE:

python3 {baseDir}/scripts/plot_data.py histogram --data samples.csv --x measurement --bins 30 --kde --title "Distribution"

Publication-quality figure:

python3 {baseDir}/scripts/plot_data.py scatter --data results.csv --x x --y y --figsize 8,6 --dpi 300 --format svg --style white --output figure1.svg

Color Palettes

• deep: Default seaborn palette
• muted: Muted colors
• bright: Bright colors
• pastel: Pastel colors
• dark: Dark colors
• colorblind: Colorblind-friendly
• viridis: Perceptually uniform
• plasma: Perceptually uniform
• RdBu: Red-Blue diverging
• coolwarm: Cool-Warm diverging

Notes

• Data can be provided as CSV files or JSON strings
• SVG output is recommended for publications
• Use --dpi 300 for high-resolution figures
• Column names with spaces should be quoted