2D Figure Generation Details

The make_figure.py script is a command-line tool for automatically generating publication-quality figures from CSV files generated by LNB-MDT analysis modules.

Overview

This tool automatically:

Detects the data type from your CSV file
Determines available plot types (line, bar, scatter, heatmap)
Generates high-resolution figures (300 DPI) suitable for publications

Prerequisites

Python 3.x with required packages (matplotlib, pandas, seaborn)
CSV files generated by LNB-MDT analysis modules

Basic Usage

Simple Command

LNB-MDT FIGURE -f <input_csv> -o <output_path>

Example:

LNB-MDT FIGURE -f cases_lnb/area_results.csv -o cases_lnb/area_plot.png

Interactive Mode (Default)

When you don’t specify --plot-type, the tool will display an interactive menu:

LNB-MDT FIGURE -f cases_lnb/height_results.csv -o cases_lnb/height_plot.png

Output:

Reading file: cases_lnb/height_results.csv
Data description: Height (nm)
Data shape: (150, 25)
Time unit: ns

Available plot types:
  1. Line Chart
  2. Bar Chart
  3. Scatter Chart
  4. Generate All

Please select plot type (1-4):

Non-Interactive Mode

Use --no-interactive flag to skip the menu and use the default plot type:

LNB-MDT FIGURE -f cases_lnb/area_results.csv -o cases_lnb/area_plot.png --no-interactive

Specify Plot Type

Directly specify the plot type:

LNB-MDT FIGURE -f cases_lnb/cluster_results.csv -o cases_lnb/cluster_plot.png --plot-type line

Command-Line Options

Command-Line Options
Option	Short	Description	Required
`--file`	`-f`	Input CSV file path	Yes
`--output`	`-o`	Output image file path	Yes
`--plot-type`		Plot type: `line`, `bar`, `scatter`, `heatmap`, or `all`	No
`--no-interactive`		Non-interactive mode (use default if plot-type not specified)	No

Supported Plot Types

The tool automatically detects available plot types based on your CSV file content:

Line Chart (`line`)

Supported for: All data types
Use case: Time series data, trends over time
Example:

LNB-MDT FIGURE -f cases_lnb/area_results.csv -o cases_lnb/area_line.png --plot-type line

Bar Chart (`bar`)

Supported for: LipidsFigure, BubbleFigure, DensityTime
Use case: Comparing values across different categories
Example:

LNB-MDT FIGURE -f cases_lnb/anisotropy_results.csv -o cases_lnb/anisotropy_bar.png --plot-type bar

Scatter Chart (`scatter`)

Supported for: LipidsFigure only
Use case: Relationship between variables, distribution analysis
Example:

LNB-MDT FIGURE -f cases_lnb/sz_results.csv -o cases_lnb/sz_scatter.png --plot-type scatter

Heatmap (`heatmap`)

Supported for: DensityFigure (Radius method) only
Use case: Multi-dimensional density data visualization
Example:

LNB-MDT FIGURE -f cases_lnb/density_radius_results.csv -o cases_lnb/density_heatmap.png --plot-type heatmap

Generate All (`all`)

Supported for: All data types
Use case: Generate all available plot types at once
Example:

LNB-MDT FIGURE -f cases_lnb/area_results.csv -o cases_lnb/area_plots --plot-type all

When using all, the tool will generate multiple files with suffixes: - area_plots_line.png - area_plots_bar.png - area_plots_scatter.png (if supported)

Data Type Detection

The tool automatically detects the data type from your CSV file:

LipidsFigure (Type 0)

Detected by: Contains Resname column
Supported plots: Line, Bar, Scatter
Examples: Area, Height, Sz Order Parameter, Mean Curvature

BubbleFigure (Type 1)

Detected by: No Resname column, bubble-related data
Supported plots: Line, Bar
Examples: Anisotropy, Gyration, Cluster, Density (single radius)

DensityFigure (Type 3)

Detected by: Description contains “DensityTime”, “Density Radius”, “Multi-Radius Density”, etc.
Supported plots: - DensityTime: Line, Bar - DensityRadius: Line, Heatmap
Examples: DensityTime, DensityRadius, Multi-Radius Density Analysis

Examples

Example 1: Generate Line Chart for Area Analysis

LNB-MDT FIGURE \
    -f cases_lnb/area_results.csv \
    -o cases_lnb/area_line.png \
    --plot-type line

Example 2: Generate All Plot Types for Height Analysis

LNB-MDT FIGURE \
    -f cases_lnb/height_results.csv \
    -o cases_lnb/height_plots \
    --plot-type all

This will generate: - height_plots_line.png - height_plots_bar.png - height_plots_scatter.png

Example 3: Generate Heatmap for Density Analysis

LNB-MDT FIGURE \
    -f cases_lnb/density_radius_results.csv \
    -o cases_lnb/density_heatmap.png \
    --plot-type heatmap

Example 4: Batch Processing (Non-Interactive)

# Generate figures for multiple analysis results
for csv_file in cases_lnb/*_results.csv; do
    base_name=$(basename "$csv_file" _results.csv)
    LNB-MDT FIGURE \
        -f "$csv_file" \
        -o "cases_lnb/figures/${base_name}_plot.png" \
        --plot-type line \
        --no-interactive
done

Example 5: Generate All Available Types Interactively

LNB-MDT FIGURE \
    -f cases_lnb/cluster_results.csv \
    -o cases_lnb/cluster_plot.png

The tool will show available options and let you choose.

Output Files

Format: PNG images
Resolution: 300 DPI (publication quality)
Size: Automatically optimized with bbox_inches='tight'
Naming: - Single plot: Uses the specified output path - Multiple plots (all): Adds suffix _<plot_type>.png

CSV File Format

The tool expects CSV files generated by LNB-MDT analysis modules with the following structure:

Standard Format

Description,Frame,Time(ns),Resname,0.0,1.0,2.0,...
Area(nm^2),0,0.0,DPPC,0.65,0.67,0.69,...
Area(nm^2),1,0.0,DAPC,0.72,0.74,0.71,...

Key Columns

Description: Analysis type (e.g., “Area(nm^2)”, “Height (nm)”)
Frame: Frame number
Time(ns): Simulation time in nanoseconds
Resname: Residue name (for LipidsFigure)
Numeric columns: Time points or other dimensions

Supported Analysis Types

The tool supports figures for all LNB-MDT analysis modules:

Supported Analysis Types
Analysis Module	CSV File	Supported Plot Types
Area	`area_results.csv`	Line, Bar, Scatter
Height	`height_results.csv`	Line, Bar, Scatter
Cluster	`cluster_results.csv`	Line, Bar
Anisotropy	`anisotropy_results.csv`	Line, Bar
Gyration	`gyration_results.csv`	Line, Bar
Sz Order Parameter	`sz_results.csv`	Line, Bar, Scatter
Density (Time)	`density_time_results.csv`	Line, Bar
Density (Radius)	`density_radius_results.csv`	Line, Heatmap

Integration with Analysis Workflow

Typical workflow:

# 1. Run analysis
LNB-MDT AREA \
    -g cases_lnb/lnb.gro \
    -x cases_lnb/lnb.xtc \
    -o cases_lnb/area_results.csv \
    -r "{'DPPC': ['PO4'], 'DAPC': ['PO4'], 'CHOL': ['ROH']}"

# 2. Generate figure
LNB-MDT FIGURE \
    -f cases_lnb/area_results.csv \
    -o cases_lnb/area_plot.png \
    --plot-type line

Troubleshooting

Error: “Input file does not exist”

Solution: Check the file path. Use relative paths from the project root directory.

Error: “Failed to read CSV file”

Solution: Ensure the CSV file is properly formatted and generated by LNB-MDT analysis modules.

Error: “Unable to determine available plot types”

Solution: The CSV file may not match expected formats. Check that it was generated by an LNB-MDT analysis module.

Error: “Plot type ‘X’ is not available”

Solution: The specified plot type is not supported for this data type. Use --plot-type all to see available options.

Warning: “Plot type ‘X’ is not supported, skipping”

Solution: This is normal when using --plot-type all. The tool will skip unsupported plot types and generate the rest.

Figures not displaying correctly

Solution: Ensure matplotlib backend is set correctly. The tool uses ‘Agg’ backend for non-interactive mode.

Tips

Use ``–plot-type all`` to see all available visualization options for your data
Check output directory exists before running (or the tool will create it)
Use descriptive output names to easily identify generated figures
Batch processing: Combine with shell scripts for processing multiple CSV files
Publication quality: All figures are generated at 300 DPI by default

Notes

The tool automatically detects the appropriate figure class (LipidsFigure, BubbleFigure, or DensityFigure) based on CSV content
Output images are saved in PNG format with 300 DPI resolution
The tool uses non-interactive matplotlib backend (‘Agg’) to avoid GUI dependencies
All paths are relative to the project root directory