Just as a machine is built from various components, the brain is composed of nearly thousands of different cell types. The differences between these cell types are partly determined by variations in their gene expression, which are regulated through transcriptional control—a process involving the dynamic activation of thousands of genes within individual cells (dynamics) and alternative splicing to generate diverse isoforms (diversification).

The regulation of RNA complexity is partly controlled by non-coding regions of the genome, known as cis-regulatory elements (CREs). These regions are bound by proteins that influence the transcription of nearby and sometime distant genes. In our previous work, we developed next-generation sequencing (NGS)-based approaches—from CRISPR screening (Diao*, Fang* et al., Nat. Methods) and 3D genome sequencing (Fang*, Yu* et al., Cell Res., 2016) to single-cell epigenomics (Preissl*, Fang* et al., Nat. Neuro., 2018) —to investigate how CREs regulate gene expression in primary cells and in the mammalian brain. We also develop bioinformatics softwares (i.e., SnapATAC) to facility the analysis of large-scale single cell epigenomic datasets (Fang et al., Nat. Commun., 2021).

We will continue developing genomic tools to understand, at the single-cell level, how the activation of specific genes and their transcriptional dynamics are regulated by CREs as well as other genetic elements across various brain cell types, how alternative splicing is affected by the experience of cells, how these experience-dependent gene expression changes shape cellular phenotypes both in vitro and in vivo, and how these genetic circuits go awry in the cases of disease.

Just as a machine relies on the precise operation of its mechanical components, the brain depends on the coordinated function of its building blocks—cells. Brain cells are interconnected by networks of genes that synchronize brain function, and when their communication breaks down, diseases can arise. Therefore, accurately identifying the types of cells involved (who), the nature of their interactions (what), their locations (where), and the timing of these interactions (when) is essential to understanding the cellular and molecular biology of multicellular systems.

We have previously developed spatially resolved transcriptomics, a method that directly visualizes the expression of thousands of RNA species in tissue sections of hundreds of micrometers thickness, enabling the study of cell-cell interactions in complex tissue at high resolution (Fang*, Xia* et al., Science, 2022; Fang*, Halpern* et al., eLife, 2023).

Our lab will continue advancing genome-scale spatial transcriptomic methods to map, at the tissue level, how cells ‘talk’ to one another in the brain. Using these data, we aim to develop predictive AI models to identify the causal genes and signaling pathways that drive these cellular interactions. Ultimately, we seek to leverage this knowledge to perturb genes or design drugs that can modify cell communication, with the goal of addressing diseases affecting the human brain.

Our genome influences nearly every aspect of our lives—from molecular functions to cellular behavior in both health and disease. Studying differences in DNA sequences between individuals could lead to the discovery of previously unknown biological mechanisms and guide the development of new therapeutic agents. Like a dynamic machine operating in real time, all living cells constantly adapt to their local environment and respond to changes through dynamic behaviors. However, linking genomic variations to diverse cellular behaviors—such as cell migration, neuronal activity and immune response—remains challenging and slow. Moreover, how most genetic associations with common diseases disrupt cellular behaviors is still largely unknown. New approaches are needed to accelerate research and unlock the vast untapped potential for understanding brain function and improving health.

To uncover these insights, we aim to combine live-cell imaging with functional genomics to comprehensively map the molecular and cellular effects of genomic variants. We will create maps across various cell types to show how coding and noncoding variants alter gene regulation and how these changes influence cellular behavior and dynamics through gene-regulatory networks. This combination of experimental data and computational predictions will help us explore how genomic variations impact brain function and disease.