As the key organ that separates human from other non-human primates, brain has continuously evolved to adapt to the changes of environments and climates. Although human shares most genetic, molecular and cellular features with primates like macaque, there are significant differences in the structure and function in brain of human and these species. Thus, exploring the differences between brains of human and non-human primates in the context of evolution will provide insights into the development, function and diseases of human nervous system. Since the genes involved in many aspects of human brain are under common pressures of natural selection, their evolutionary features can be analyzed collectively at pathway-level. In this study, we tried to explore the molecular mechanisms underlying human brain by comparing the evolutionary features of pathways enriched in genes expressed in human brain and macaque brain. We identified 31 differentially evolutionary pathways in the brain of human and macaque, among which included those related to neurological diseases, signaling transduction, immune and metabolism. By analyzing differentially expressed genes in brain regions and development stages between human and macaque, we also found that ten and four pathways with evolutionary differences, respectively. We further performed crosstalk analysis of pathways to obtain an intuitive correlation between the pathways, which is helpful to understand the mechanism of interaction between pathways. Our results shed light on a comprehensive view of the evolutionary pathways of the human nervous system and provide a reference for the study of human brain development.

This is a supplemtary figure showing the distribution of variance component corresponding to dN/dS or pN/pS of real and simulated pathways. Variance component for pathways was computed in the following way, i.e., 30 pathways (~10% of the 313 pathways analyzed) were randomly selected from the pathway set and the variance component due to differences between pathways of dN/dS (or pN/pS) was computed. Then, genes in the selected pathways were reshuffled and variance component corresponding to the simulated pathway sets were calculated based on the permuted dN/dS (or pN/pS) ratios. The procedure was repeated 5,000 times to construct the distribution of variance component of pathway set and the null distribution. If the genes in the same pathway do not share the same evolutionary properties, the distribution of the variance component due to differences between pathways should be similar to the null distribution. In each panel, the variance component of null distribution and real pathways are shown in dark and gray, respectively.