Skip to content

Junyue Cao (keynote ★) — scalable combinatorial-indexing single-cell, plus temporal and spatial dynamics

Speaker: Junyue Cao, Rockefeller University (confirmed — SCGD26 speaker page). (Trained as an early student in the Shendure Lab, UW; colleague Darren (Cusanovich) on sci-ATAC-seq.) Slot: SCGD26, Jun 12, 2026 — 3:40–4:20 PM EDT (40-min featured talk), "Scalable genomic approaches to investigate cellular dynamics in aging and disease" Recording: SCGD26 livestream (YouTube) — verify archived link Status: Captured from session transcript

Thesis

The body makes millions of new cells per second across thousands of cell types defined by thousands of integrated molecular features — so studying one gene or one cell misses the drivers and the cross-cell interactions. The lab builds highly scalable tools to quantify cell-population dynamics: which populations change, their temporal (proliferation/differentiation) and spatial interactions, and the molecular drivers.

Scalable single-cell (combinatorial indexing)

  • Combinatorial indexing barcodes molecules by repeated split-pool — no single-cell isolation. Lineage: sci-ATAC-seq (Darren) → sci-RNA-seq → sci-CAR (RNA+ATAC co-assay, 2018) → sci-RNA-seq3 (millions of cells, embryo).
  • EasySci (students Andras [Sziraki] + Jasper): full-gene-body capture (exon-level → splice/isoform + cell-state) at ~$700 library prep / 1M cells. Used across the brain (~2M cells, rare neurons) and a whole-organism atlas (>20 tissues, >1,000 tissues / >30M cells). Quantifies per-cell-type dynamics across aging (e.g. GZMK⁺ CD8 T-cell female-specific expansion; Th17 male-specific expansion; ~100 immune states × >20 tissues).
  • EnrichSci (bioRxiv 2025.05.02.651937; Cell Genomics S2666-979X(25)00357-X): RNA-probe enrichment against cell-type-specific gene modules (program-level, not single-gene/antibody) to target rare types — one module dimension separates/enriches 4 cell types in one experiment.

Temporal dynamics (proliferation + differentiation)

  • sci-fate (heard as "SkyFate"; Cao et al., Nat. Biotechnol. 38, 980–988, 2020): scRNA + 4sU metabolic labeling → two transcriptomes (newly synthesized + pre-labeling) per cell → infer cell-state transition trajectories and probabilities from one experiment; reconstructs cell-cycle transitions and predicts population dynamics over hours.
  • TrackerSci ("Tracking cell-type-specific temporal dynamics in human and mouse brains," Cell, 2023): EdU labeling of newborn cells + scRNA/ATAC → quantify proliferation and differentiation rates in vivo across >20 tissues (>2M newborn cells; e.g. reduced neural-progenitor proliferation + increased microglial proliferation in aged brain; reduced OPC differentiation), mapping the gene/chromatin regulators of turnover.

Spatial dynamics (sequencing, not imaging)

  • IRISeq (image reconstruction by sequencing; "Optics-free Spatial Genomics for Mapping Mouse Brain Aging," bioRxiv 2024.08.06.606712 / PMC11326199): instead of scanning each cell's location, every bead records its neighbors (Hi-C-like), so location is reconstructed from local neighborhoods at constant time regardless of area. (Concurrent with related approaches — e.g. SCOPE and other imaging-free spatial genomics.)
  • Implementation (first author Abdul [Abdulraouf Abdulraouf]; senior author Wei Zhou): tens of millions of in-house sender + receiver beads, combinatorially barcoded; receiver beads capture nearby tissue mRNA (polyT) + photocleaved sender-bead signal → reconstruct bead positions + cluster into regions. Validated on mouse brain (30 regions), 5–50 µm beads, ~$30/section (<$1/mm²); mapped region-specific cell-population changes (and interactions) in adult (4 mo) vs. aged (23 mo) brain.
  • PerturbFate (grad student Zihan Xu; Rockefeller news, 2026; builds on PerturbSci-Kinetics, Xu et al., Nat. Biotechnol. 2023): couples high-throughput perturbation with the scalable single-cell + multi-layer (4sU nascent-RNA + ATAC) read-out to find molecular/cellular regulators of cellular alterations.

Connections to the corpus

Verify / open questions

  • Confirmed: speaker = Junyue Cao (Rockefeller), talk title, and slot (SCGD26 page). Method names corrected/confirmed: EasySci, EnrichSci, sci-fate (was "SkyFate"), TrackerSci, PerturbFate, IRISeq (the spatial image-reconstruction method) — see Sources.
  • Still unverified: archived recording URL; student "Jasper" surname; whether "EasySci" inline DOI should be the brain-aging atlas paper vs. the method protocol (not pinned here).

Sources

  • SCGD26 speaker/schedule page: https://satijalab.org/scgd26/
  • Junyue Cao lab (Rockefeller): https://www.rockefeller.edu/our-scientists/heads-of-laboratories/8704-junyue-cao/
  • EnrichSci — bioRxiv: https://www.biorxiv.org/content/10.1101/2025.05.02.651937v1 ; Cell Genomics: https://www.cell.com/cell-genomics/fulltext/S2666-979X(25)00357-X
  • sci-fate — Nat. Biotechnol. 2020: https://www.nature.com/articles/s41587-020-0480-9
  • TrackerSci — Cell 2023: https://www.sciencedirect.com/science/article/pii/S0092867423009765
  • IRISeq — bioRxiv 2024.08.06.606712 / PMC11326199: https://pmc.ncbi.nlm.nih.gov/articles/PMC11326199/
  • PerturbFate — Rockefeller news 2026: https://www.rockefeller.edu/news/39376-novel-tool-could-identify-new-therapeutic-targets-in-complex-diseases-like-cancer/ ; PerturbSci-Kinetics — Nat. Biotechnol. 2023: https://www.nature.com/articles/s41587-023-01948-9