Tooluniverse Image Analysis

What it does

This skill, Tooluniverse Image Analysis, enables AI agents to analyze microscopy-derived measurement data and perform quantitative imaging analysis. It utilizes libraries like pandas, numpy, scipy, statsmodels, and scikit-image to process data from image analysis software such as ImageJ, CellProfiler, or QuPath. The skill prioritizes a data-first approach, emphasizing loading and inspecting data before performing statistical analyses.

When to use it

This skill is useful for:

• Analyzing microscopy measurement data (area, circularity, intensity, cell counts) stored in CSV/TSV files.
• Performing colony morphometry or cell counting statistics.
• Conducting fluorescence quantification and statistical comparisons like t-tests, ANOVA, Dunnett's tests, Mann-Whitney tests, and power analysis.
• Building regression models (polynomial, spline) for dose-response or ratio data.

Key capabilities

• Loads and inspects CSV/TSV microscopy measurement data.
• Performs statistical comparisons (t-tests, ANOVA, Dunnett's test, Mann-Whitney).
• Calculates effect sizes (Cohen’s d) and performs power analysis.
• Builds regression models (polynomial, spline) for dose-response or ratio optimization.
• Understands common ImageJ/CellProfiler columns like Area, Circularity, Round, and Intensity.

Example prompts

• "What is the mean circularity of genotype X?"
• "Perform a t-test comparing NeuN counts between condition A and condition B."
• "Build a polynomial regression model for this dose-response data."

Tips & gotchas

• The skill requires microscopy measurement data in CSV/TSV format. It is not suitable for raw image files (TIFF, PNG) directly – those require segmentation and measurement steps first.
• It's designed specifically for imaging context; avoid using it for statistics without a connection to image analysis.
• Always load and inspect the data before requesting specific analyses to ensure accuracy and understanding of the dataset.