CosinorPy

CosinorPy implements cosinor-based statistical methods to detect and analyze rhythmic (periodic) patterns in biological time-series data.

Key Features:

• Model Flexibility: Supports single- and multi-component cosinor models for capturing multiple rhythmic components within time-series data.
• Automatic Model Selection: Provides automatic model selection to identify the most appropriate cosinor model for a given dataset.
• Population-Mean Cosinor Regression: Implements population-mean cosinor regression for group-level analysis of rhythmicity.
• Differential Rhythmicity Assessment: Provides methods to assess differential rhythmicity between conditions or groups.
• Experimental Design Support: Includes functions to assist experimental design optimized for subsequent rhythmic analysis.
• Synthetic Data Generation: Generates synthetic datasets for hypothesis testing and method validation.
• Data Import/Export Flexibility: Supports import and export of data in various formats to accommodate different input types.

Scientific Applications:

• Chronobiology: Analysis of circadian and other biological rhythms in time-series datasets.
• Circadian Gene Expression: Detection and characterization of periodic patterns in gene expression data.
• Hormone and Physiological Profiling: Analysis of circadian patterns in hormone levels and other physiological processes.

Methodology:

Computational methods include classical trigonometric regression based on cosinor methods, using single- and multi-component cosinor models, population-mean cosinor regression, and automatic model selection.