Research: Neurogenomics

1. Gene Regulation and Neuroplasticity: 

We pursue biology-driven discovery. Using advanced epigenomic and genomic sequencing approaches (e.g., single-cell omics, spatial omics), we dissect activity-dependent gene-regulatory programs in memory formation and stress responses in vivo. We then test causality through gene knockdown/overexpression, optogenetics, and circuit-based perturbations.

In a recent study (Itoh, Khalil, et al., bioRxiv 2026), we explored how experience reshapes the brain at the level of neuronal ensembles, chromatin regulation, and gene expression. We used and developed single-cell multiomics technologies to profile transcription, chromatin accessibility, histone modifications (H3K27ac, H3K27me3), and transcription factor (FOS) activity in memory engram cells. In parallel, we built computational methods such as ChromTRAP to identify recently activated neuronal populations from epigenomic traces and uncover the molecular programs that link neural activity to long-term plasticity. Through this integrated experimental and computational approach, we revealed how memories and other experience-dependent brain states are encoded in the genome regulatory landscape of neurons. 

2. Development of Molecular Recording Technologies:

Cells display profound diversity shaped by layered transcriptional and epigenetic regulatory programs during development, learning, aging, and disease. Single-cell sequencing has revealed the molecular heterogeneity and regulatory dynamics underlying these processes in complex tissues. However, because these methods require cell lysis to harvest nucleic acids, they capture only static snapshots, after which no further observations are possible. Thus, even when sampling multiple time points, it remains challenging to unambiguously trace the molecular trajectories of individual cells. 

We recently developed HisTrac-seq, a "time-machine"-like whole-genome history-tracing platform that records transcriptional and epigenetic states in genomic DNA (Kawamura, Khalil et al., bioRxiv 2025). HisTrac-seq transiently activates DNA adenine methyltransferase fused to proteins of interest (POIs), thereby installing N6-methyladenine bookmarks at POI-associated genomic regions. By changing the POI, HisTrac-seq enables historical recording of diverse regulatory modalities, including transcription-associated activity, chromatin accessibility, and histone modifications. Coupled with single-cell multi-omics readout, HisTrac-seq simultaneously captures two temporally distinct molecular layers in the same single cells, pairing past and present gene regulatory states. 

We are now extending this platform toward spatiotemporal history tracing and integration with physiological readouts, while actively disseminating these tools.