Chart By Numeric X
Visualize numeric relationships with scatter plots, trend lines, and R-squared correlation analysis for section-specific data exploration
Specs
Version
0.1.0 (updated on 2025-04-20)
Developer
Labii Inc.
Type
Section
Support Configuration
Yes
Overview
The Chart By Numeric X widget creates interactive scatter plots and line graphs to visualize relationships between numeric variables within your experimental records. By plotting numeric values on both x and y axes, this widget enables researchers to identify correlations, trends, and patterns in their data. With built-in R-squared prediction capabilities, the widget can generate best-fit lines with regression formulas to quantify relationships between variables. This powerful analytical tool is essential for dose-response curves, calibration standards, time-series analysis, and any scenario where understanding numeric correlations supports scientific interpretation and decision-making.
Use Cases
Dose-Response Curves: Plot drug concentration versus biological response to determine EC50 values and therapeutic windows
Calibration Standards: Create standard curves for analytical methods by plotting known concentrations against instrument readings
Time-Series Analysis: Visualize how measurements change over time to identify trends, cycles, or anomalies
Correlation Studies: Explore relationships between two numeric variables (temperature vs. reaction rate, cell density vs. protein expression)
Quality Control: Plot measurement precision data to assess instrument performance and method reliability
Growth Curves: Track cell proliferation, bacterial growth, or biomass accumulation over time
Kinetic Analysis: Visualize enzyme kinetics, reaction rates, or binding affinities with velocity versus substrate concentration plots
Performance Validation: Compare expected versus observed values to validate experimental methods and models
Interface
Read-only View
The read-only view displays an interactive scatter plot with numeric values on both axes. Users can hover over data points to view exact values and switch between multiple chart types for optimal data visualization.
Data Display: Scatter plot showing individual data points with numeric x and y coordinates
Interactive Features: Hover over points to view exact values, coordinates, and data labels
Chart Type Selection: Switch between scatter, line, area, and scatter-with-fit visualizations
Regression Analysis: "Scatter with fit" option displays best-fit line with R-squared value and regression equation
Export Access: Download button for extracting source data for external analysis
R-squared (R²): The coefficient of determination represents the proportion of variation in the dependent variable (y-axis) that can be predicted from the independent variable (x-axis). Values closer to 1 indicate stronger correlations. The "Scatter with fit" tool generates the best-fit line and displays the regression formula on screen.
Edit View
The edit view is not applicable for this widget as it is designed purely for data visualization and analysis. All configuration and customization is performed through the widget settings panel accessible via the configure button in the widget header. The widget automatically generates visualizations based on the configured parameters without requiring direct editing of the chart.
Configuration
Initial Setup
Add the Chart By Numeric X widget to your section by clicking Add Widget and selecting Chart By Numeric X from the Data Driven Charts category
Click the Configure button or settings icon in the widget header to open the configuration panel
Configure the data source and axis parameters as described below
Required Settings
These settings must be configured for the widget to function properly:
Table: Select the table that contains the numeric data you want to visualize. This selection determines which columns are available for the x-axis and series configuration
X axis: Choose a column for the x-axis. This column must contain numeric data using Number, Formula, or Consumption widgets. The x-axis typically represents the independent variable in your analysis
Series: Choose one or more columns for the y-axis data series. These columns must also contain numeric data using Number, Formula, or Consumption widgets. The series represent the dependent variables you're analyzing
The widget requires columns with Number, Formula, or Consumption widgets that return numeric values for both the x-axis and series. Text, date, or categorical columns cannot be used with this widget.
Optional Settings
Customize your visualization with these optional parameters:
Data Filtering
Filter: Enter a filter expression to limit which records are included in the chart. Leave empty to include all records from the selected table. Filters help focus the visualization on specific subsets of data (e.g., specific experimental conditions, date ranges, or quality criteria)
Visualization Customization
Chart title: Provide a descriptive title for your chart. This title appears at the top of the visualization and should clearly indicate what relationship or data is being displayed
Height: Set the chart height in pixels. Default is 250px. Adjust based on the density of your data points and available section space. Larger heights provide better readability for complex datasets
Select your Table to define the data source
Choose an X axis column with numeric data (independent variable)
Select one or more Series columns with numeric data (dependent variables)
Optionally add a Filter to limit which records are included
Provide a descriptive Chart title and adjust Height as needed
Click Save to apply your configuration and generate the chart
Advanced Configuration
For complex analytical scenarios:
Use Filter expressions to compare different experimental conditions by creating multiple widget instances with different filters
Configure multiple Series columns to overlay several dependent variables on the same x-axis for comparative analysis
Adjust Height to accommodate dense data sets or multiple series without visual clutter
Use the scatter-with-fit chart type to quantify relationships and extract regression equations for further calculations
When plotting multiple series with very different value ranges, consider creating separate charts or normalizing data to prevent one series from dominating the visualization scale.
Additional Functions
Chart Type Selection
The Chart By Numeric X widget supports multiple visualization types to best represent your numeric relationships:
Available Chart Types:
Scatter: Individual data points without connecting lines, ideal for showing distributions and identifying outliers
Line: Connected data points showing progression and trends, best for time-series or ordered data
Area: Filled area under the line emphasizing magnitude and cumulative values
Scatter with Fit: Scatter plot with best-fit regression line, R-squared value, and equation displayed
After the chart is generated, locate the chart type selector in the widget interface
Click on different chart type options to switch visualization styles
Select "Scatter with fit" to enable R-squared regression analysis
The chart updates immediately to reflect the selected visualization type
Use scatter plots for exploratory analysis to identify patterns, then switch to scatter-with-fit to quantify relationships when correlations are apparent.
R-squared Prediction and Regression Analysis
The scatter-with-fit option provides powerful statistical analysis for understanding relationships between numeric variables:
What You Get:
Best-fit Line: Visual representation of the linear regression through your data points
Regression Equation: Mathematical formula (y = mx + b) describing the relationship
R-squared Value: Statistical measure (0-1) indicating how well the line fits the data
Slope (m): Rate of change in y for each unit change in x
Intercept (b): Predicted y-value when x equals zero
Ensure your chart is displaying data with a clear numeric relationship
Select the Scatter with fit chart type
The widget automatically calculates and displays the best-fit line through your data
View the regression equation and R-squared value displayed on the chart
Use the equation to predict values or the R-squared to assess correlation strength
Interpreting R-squared Values:
R² ≥ 0.9: Strong correlation, model fits data very well
0.7 ≤ R² < 0.9: Good correlation, reasonable predictive value
0.5 ≤ R² < 0.7: Moderate correlation, some predictive utility
R² < 0.5: Weak correlation, limited predictive value
R-squared analysis assumes a linear relationship. For non-linear relationships, the fit may be poor even when a clear pattern exists. Consider data transformation or alternative analysis methods for non-linear data.
Data Export
Export the underlying numeric data used to generate your chart:
Locate the Download button in the widget interface (typically in the widget header)
Click Download to export the chart data
Save the downloaded file to your desired location
The exported data includes:
X-axis values for each data point
Y-axis values for all series
Record identifiers linking data points to source records
Filtered data only (if filters are applied)
Use exported data for:
Advanced statistical analysis in R, Python, or GraphPad Prism
Creating publication-quality figures in specialized software
Performing non-linear regression or other advanced modeling
Archiving analytical results for compliance documentation
Sharing raw data with collaborators
Multi-Series Visualization
When multiple series are configured, the chart displays all dependent variables on the same x-axis:
Color Coding: Each series is assigned a distinct color for easy identification
Legend: Automatic legend shows series names and colors
Hover Information: Hovering over points shows which series the data point belongs to
Comparative Analysis: Overlay multiple measurements to identify relationships between different dependent variables
This is particularly useful for:
Comparing replicates or technical repeats
Visualizing multiple analytes measured under the same conditions
Displaying different response measurements for the same experimental treatment
Best Practices
Data Organization
Numeric Columns Only: Ensure both x-axis and series columns contain purely numeric data from Number, Formula, or Consumption widgets
Appropriate X-axis Selection: Choose the independent variable (what you control or measure first) for the x-axis
Meaningful Series: Select dependent variables (what you measure as a result) for the series
Data Quality: Clean data by removing null values, outliers, or obvious errors before visualization to ensure accurate regression analysis
Performance Optimization
Filter Strategically: Use filters to limit data points to reasonable numbers (typically under 1000 points per series for optimal performance)
Height Adjustment: Set appropriate heights based on data density – more points may need larger charts for clarity
Limited Series: Avoid plotting more than 5-7 series on a single chart to maintain readability
Incremental Analysis: Start with smaller filtered datasets before expanding to full data ranges
Analytical Strategy
Exploratory First: Begin with basic scatter plots to identify patterns before applying regression analysis
Check Linearity: Use R-squared analysis only when relationships appear linear; non-linear data needs different approaches
Validate Outliers: Investigate data points that fall far from the regression line – they may represent errors or important discoveries
Replicate Analysis: When possible, include multiple replicates and assess consistency across series
For calibration curves, plot standard concentrations on the x-axis and instrument response on the y-axis. Use scatter-with-fit to obtain the equation, then use it to calculate unknown concentrations from their measured responses.
Common Pitfalls to Avoid
Avoid: Using categorical or text columns for axes – this widget requires numeric data
Instead: Use Chart By Category X for categorical data or convert categories to numeric codes if appropriate
Avoid: Assuming high R-squared always means causation or that the relationship is meaningful
Instead: Consider biological or chemical plausibility and validate with additional experiments
Avoid: Applying linear regression to obviously non-linear relationships
Instead: Transform data (log, sqrt, etc.) or use specialized non-linear analysis tools
Avoid: Plotting too many series with vastly different scales on one chart
Instead: Create separate charts or normalize data to similar ranges
Visualization Best Practices
Descriptive Titles: Include variable names and units in chart titles (e.g., "Cell Viability vs. Drug Concentration (μM)")
Appropriate Chart Types: Use scatter for general relationships, line for time-series, scatter-with-fit for quantifying correlations
Scale Considerations: Ensure both axes use appropriate scales that don't distort the apparent relationship
Legend Clarity: When using multiple series, ensure series names clearly identify what each represents
Statistical Considerations
Sample Size: Regression analysis is more reliable with larger numbers of data points (typically n ≥ 10)
Range Coverage: Ensure x-axis values span the range of interest; don't extrapolate far beyond measured data
Residual Analysis: Data points should distribute randomly around the regression line; patterns in residuals suggest non-linearity
Uncertainty: Remember that R-squared doesn't account for measurement uncertainty or biological variability
Compliance and Documentation
Export Regularly: Download chart data for audit trails and to document analytical methods
Document Configuration: Record filter criteria, axis selections, and regression equations in your experimental documentation
Validation: For regulatory contexts, validate that regression parameters meet acceptance criteria (e.g., R² ≥ 0.995 for analytical curves)
Standardization: Establish standard configurations for recurring analyses to ensure consistency
Related Widgets
Chart By Category X: Use together when you need both categorical and numeric visualizations. Chart By Category X handles discrete groups while Chart By Numeric X displays continuous numeric relationships
Data Visualizer: Combine with Data Visualizer to supplement table-driven numeric charts with manually entered calibration or reference data
Record Summary: Pair with Record Summary to provide high-level record patterns alongside detailed numeric relationship analysis
Integration Scenarios
Combine Chart By Numeric X + Number columns + Time tracking for time-series analysis of experimental progress
Use Chart By Numeric X + Formula columns + Calculated values to visualize derived relationships and ratios
Integrate Chart By Numeric X + Consumption widgets + Inventory data for usage rate analysis and forecasting
Pair Chart By Numeric X + Filter widgets + Treatment groups to overlay dose-response curves from multiple experiments
Combine Chart By Numeric X + Scatter-with-fit + Standard curves for quantitative analytical method validation
Last updated