Detecting change in biological rhythms: a multivariate permutation test approach to Fourier-transformed data

Area of Science:

• Neuroscience
• Chronobiology
• Biostatistics

Background:

• Analyzing neurobiological rhythms is crucial for understanding treatment effects in psychology and psychiatry.
• Existing methods often oversimplify complex biological time series data.
• Novel analytical approaches are needed to capture the intricacies of biological rhythms.

Purpose of the Study:

• To introduce and validate a statistically powerful method for analyzing changes in biological rhythms.
• To address limitations of current methods in handling complex time series data.
• To demonstrate the utility of Fourier methods combined with multivariate permutation testing (MPT) for biological rhythm analysis.

Main Methods:

• Utilized Fourier methods for transforming time series data.
• Applied multivariate permutation testing (MPT) to analyze Fourier-transformed data.
• Conducted Monte Carlo simulations to compare MPT with Bonferroni correction for statistical power and family-wise error control.
• Applied the developed method to real-world human neurotransmitter data.

Main Results:

• Multivariate permutation testing (MPT) demonstrated greater statistical power compared to Bonferroni-corrected methods.
• MPT effectively controlled family-wise error rates.
• The Fourier with MPT approach proved effective on real human neurotransmitter time series data.

Conclusions:

• Fourier methods combined with MPT offer a statistically robust and powerful approach for detecting changes in biological rhythms.
• This method enhances the analysis of complex time series data in neurobiology and chronobiology.
• The validated method provides researchers with a valuable tool for studying treatment-related rhythm alterations.