What libraries are required for these data science patterns?

The skill primarily utilizes standard Python scientific libraries including Pandas, NumPy, SciPy, Matplotlib, Seaborn, and Scikit-learn.

Does it help with identifying data outliers?

Yes, it includes patterns for visual assessment using Box Plots and Violin Plots, alongside statistical distribution checks like Shapiro-Wilk and Kolmogorov-Smirnov tests.

How does this skill help in choosing the right statistical test?

The skill provides a structured decision tree that guides you to the correct test (such as t-tests, ANOVA, or Mann-Whitney U) based on whether your data is continuous or categorical and whether it meets normality assumptions.

Can I use this skill for high-dimensional data visualization?

Yes, the skill includes optimized implementations for Principal Component Analysis (PCA) and t-SNE, including data standardization and variance explanation plots.

Does it support correction for multiple hypothesis testing?

Absolutely. It provides code patterns for Bonferroni and False Discovery Rate (FDR/Benjamini-Hochberg) corrections using the statsmodels library.

Scientific Data Science

Name: Scientific Data Science
Author: justaddcoffee

byjustaddcoffee

Data Science & ML

Provides standardized statistical analysis strategies and data exploration patterns for scientific discovery and research.

About

This skill equips Claude Code with a comprehensive framework for rigorous scientific data analysis and hypothesis testing. It offers automated patterns for initial data assessment, distribution verification, and a structured decision tree for selecting appropriate statistical tests—ranging from parametric t-tests and ANOVA to non-parametric alternatives. Designed for high-integrity research, it includes implementations for dimensionality reduction (PCA, t-SNE), clustering, and advanced practices like multiple testing correction (FDR) and power analysis to ensure that data-driven insights are both statistically sound and reproducible.

Key Features

Rigorous reporting templates for statistical significance and effect sizes
Multiple testing correction methods (Bonferroni and FDR)
Automated data exploration and distribution checking patterns
Statistical decision tree for selecting appropriate hypothesis tests
Dimensionality reduction techniques including PCA and t-SNE
0 GitHub stars

Use Cases

Identifying and adjusting for confounding variables in complex observational studies
Validating statistical assumptions before performing hypothesis testing on experimental datasets
Visualizing and clustering high-dimensional biological or physical science data

About

Key Features

Rigorous reporting templates for statistical significance and effect sizes
Multiple testing correction methods (Bonferroni and FDR)
Automated data exploration and distribution checking patterns
Statistical decision tree for selecting appropriate hypothesis tests
Dimensionality reduction techniques including PCA and t-SNE
0 GitHub stars

Use Cases

Identifying and adjusting for confounding variables in complex observational studies
Validating statistical assumptions before performing hypothesis testing on experimental datasets
Visualizing and clustering high-dimensional biological or physical science data