Mapping functional brain networks from the structural connectome: Relating the series expansion and eigenmode...

Area of Science:

• Neuroscience
• Network Science
• Computational Biology

Background:

• Brain functional networks are influenced by structural connectivity but are not direct reflections.
• Existing theories, like series expansion and eigenmode approaches, attempt to explain structure-function relationships.
• A unified framework for these theories is lacking.

Purpose of the Study:

• To analyze the relationship between the series expansion and eigenmode approaches for estimating functional brain networks from structural data.
• To determine the unique and common explanatory power of each approach.
• To provide a theoretical and empirical basis for preferring one approach over the other.

Main Methods:

• Utilized linear algebra to demonstrate the mathematical relationship between the series expansion and eigenmode approaches.
• Derived explicit coefficient expressions for both approaches.
• Validated theoretical findings using empirical data from Diffusion Tensor Imaging (DTI) and functional Magnetic Resonance Imaging (fMRI).

Main Results:

• Showed that the eigenmode approach can be expressed in terms of the series expansion approach (walks on the structural network).
• Empirically verified a strong correlation between mappings derived from both approaches using DTI and fMRI data.
• Demonstrated that the eigenmode approach consistently provides a better or equal fit to functional data compared to the series expansion approach, especially when structural data contains errors.

Conclusions:

• The eigenmode approach is theoretically linked to the series expansion approach.
• The eigenmode approach offers superior or equivalent explanatory power for functional brain networks derived from structural data.
• The eigenmode approach is recommended over the series expansion approach due to its robustness and better fit, advancing the unification of structure-function relationship theories.