Application Instructions

1 Load Data

Import chromatography data files into the application for analysis. This is the foundation step where you bring your experimental data into ChromaAnalyzer. The quality and completeness of your data will directly impact all subsequent analysis steps, so it's important to ensure your data is properly formatted and contains all necessary columns before proceeding. The application will automatically detect data types and provide a summary to help you verify that the import was successful.

Supported File Formats

• CSV files (.csv)
• Excel files (.xlsx, .xls)

1. Click the "Load Data File" button in the left panel
2. Select your data file using the file browser dialog
3. Review the Data Summary table in the right panel
4. Verify column statistics and data integrity
5. Click "Next →" when satisfied with the data import

2 Data Preview & Column Selection

Define the role of each data column and configure analysis parameters. Proper column assignment is critical for accurate analysis - the application needs to understand which columns contain your batch identifiers, signal data, volume measurements, and other key variables. This step also allows you to preview your data to ensure the assignments are correct before moving forward. Getting this right will ensure all subsequent calculations and visualizations use the appropriate data.

Column Assignment

1. Use dropdown menus to assign column roles
2. Configure bed height settings as needed
3. Use the Preview Data button to verify the data
4. Click "Next →" to proceed to preprocessing

3 Preprocessing

Configure and apply data preprocessing steps to prepare for transition analysis. The level of smoothing required will depend on the level of noise in the data and resolution of the data. Care should be taken to ensure that the data is smoothed to reduce noise while also maintaining the original shape of the signal. This is particularly important for the derivative curve, as some metrics are particularly sensitive to the level of processing. This step will always require some level of subjectivity.

Available Options

1. Select desired preprocessing methods in the left panel
2. Click "Preview Preprocessing" to see effects. Verify impact across multiple batches
3. Data should be smoothed to reduce noise while also maintaining the original shape of the signal
4. Click "Next →" when satisfied with preview to apply preprocessing to the entire dataset and to calculate transitional analysis metrics

4 Results

Review calculated transition analysis metrics and export results. This step presents the core analytical outputs from your chromatography data - the calculated performance metrics that quantify column behavior and efficiency. These metrics form the foundation for all subsequent statistical analysis, control charting, and process monitoring. Take time to review the values for reasonableness and export the data for documentation or further analysis in other tools.

Available Metrics

1. Review calculated metrics in the summary table
2. Perform sense check of transition analysis metrics
3. Optionally export the metrics as a CSV file for further analysis
4. Proceed to the Control Limits step via the "Next →" button.

5 Control Limits

Calculate statistical control limits for process monitoring and outlier detection. Control limits establish the boundaries of normal process variation based on your historical data. These statistical boundaries are essential for distinguishing between common cause variation (normal process fluctuations) and special cause variation (unusual events that require investigation). By setting these limits using stable reference batches, you create a baseline for ongoing process monitoring and can objectively identify when your chromatography process is operating outside expected parameters.

Control Limit Types

1. Select reference batches using the batch selection controls to calculate control limits
2. Click "Calculate Control Limits" to compute statistical boundaries
3. Perform sense check of control limit output
4. Export control limits data if needed
5. Click "Next →" to proceed to visualization

6 Control Chart Heatmap

Visualize process performance across all metrics and batches using the heatmaps. Heatmaps provide a powerful overview of your entire dataset at once, allowing you to quickly spot patterns, correlations, and outliers across multiple metrics simultaneously. This bird's-eye view is invaluable for identifying systematic issues, understanding how different metrics relate to each other, and pinpointing batches or time periods that require investigation. The color-coding makes it easy to see which areas of your process are performing well and which need attention.

Understanding the Heatmap

The heatmap uses a colorblind friendly palette based on the number of standard deviations from the process mean. Limits are based on those calculated on the previous step. Note that in order to prevent large outliers from skewing the color mapping, the scale maxes out at greater than or equal to 3 standard deviations from the process mean.

1. Click "Generate Heatmap" to create visualization
2. Review heatmap for process patterns and outliers
3. Export heatmap image if needed
4. Click "Next →" for individual control charts

7 Control Charts

Generate individual statistical process control charts for detailed metric analysis. Control charts are essential for monitoring process stability over time and detecting when your chromatography process goes out of statistical control. They help you identify trends, shifts, or unusual patterns that might indicate equipment issues, resin degradation, or subtle process shifts.

Chart Features

• Control Limits: UCL, LCL, and center line based on Step 5
• Data Points: Individual batch results with trend lines
• Out-of-Control Highlighting: Red halo effect for problematic batches
• ID Preservation: Original sample colors maintained with alert highlighting

1. Select metric from dropdown menu in left panel
2. Configure ID-based coloring if desired
3. Click "Plot Control Chart" to generate visualization
4. Review chart for out-of-control conditions and trends
5. Export chart image if needed
6. Repeat for additional metrics as needed

8 Clustering Analysis

Perform advanced pattern recognition and batch grouping analysis. Clustering analysis helps identify hidden patterns and relationships between batches that may not be obvious from individual control charts or heatmaps. This is particularly valuable for understanding process variations, identifying similar operating conditions, detecting outlier groups, and optimizing process parameters.

Clustering Methods

1. Configure clustering parameters in the left panel or leave options as "auto" (default)
2. Select whether to use core metrics (Kim et al.) or all metrics
3. Click "Generate Clustering Analysis" to run analysis
4. Review clustering results and visualizations
5. Export clustering plots as required