Quantifying the influence of biophysical factors in shaping brain communication through remnant functional networks
Functional connectivity (FC) reflects brain wide communication crucial for cognition, yet the role of underlying biophysical factors, physical and molecular, in shaping FC remains unclear. We quantify the influence of physical factors, structural connectivity (SC) and spatial autocorrelation (DC), capturing anatomical wiring and distance between regions and molecular factors, gene expression similarity (GC) and neuroreceptor congruence (RC), capturing similarity in the neurobiological make up on the organizational features of fMRI derived resting state FC. We assess the impact of these factors on graph-theoretic and topological features, capturing both pairwise and higher order interactions between brain regions. We develop a simple test using remnant functional networks generated by selectively removing connections aligned with specific biophysical factors. Our findings reveal that molecular factors, particularly RC, predominantly shape graph-theoretic features, while topological features are influenced by a mix of molecular and physical factors, notably GC and DC. Surprisingly, SC plays a minor role in FC organization. Additionally, we link FC alterations to specific biophysical factors in neuropsychiatric conditions such as schizophrenia, bipolar disorder, and ADHD, with physical factors more effectively differentiating these groups. These insights are crucial for understanding and modeling FC across various applications. Our analysis offers a robust method to examine how underlying factors dynamically influence FC in different contexts beyond resting state, including during a task, development, and clinical conditions.