Back

Integration of in situ hybridization and scRNA-seq data provides a 2D topographical map of the developing retina across species

Joisher, H. N.; Lee, C.; Prabhakara, C.; van der Weide, I.; Si, Y.; Lonfat, N.; Cepko, C.

2026-01-04 developmental biology
10.64898/2026.01.04.697548 bioRxiv
Show abstract

Precise regional patterning is fundamental to tissue organization, yet the spatial logic that governs it remains poorly defined for many tissues. In the vertebrate retina, molecular domains along the dorsoventral and nasotemporal axes provide positional cues for regional specializations such as the high-acuity area (HAA). We combined multiplexed in situ hybridization data with single-cell transcriptomic data to create quantitative two-dimensional maps of developing retinal cells. In the developing chicken retina, this approach resolved sharp expression boundaries of genes involved in patterning, and revealed novel candidates enriched in the anlagen of the HAA. Comparative analysis of chicken, mouse, and human data demonstrated conserved axis-based programs, but distinct fine-scale organization consistent with presence/absence of an HAA. Here, we show that spatial reconstruction from scRNA-seq data, anchored by experimental benchmarks, enables comparative 2D topographic mapping of gene expression across species and provides a generalizable strategy to investigate the spatial logic of molecular organization in developing tissues. GRAPHICAL ABSTRACT O_FIG O_LINKSMALLFIG WIDTH=200 HEIGHT=141 SRC="FIGDIR/small/697548v1_ufig1.gif" ALT="Figure 1"> View larger version (44K): org.highwire.dtl.DTLVardef@e9b84aorg.highwire.dtl.DTLVardef@4207deorg.highwire.dtl.DTLVardef@112bc79org.highwire.dtl.DTLVardef@113e1d3_HPS_FORMAT_FIGEXP M_FIG C_FIG HIGHLIGHTSO_LIQuantification of multiplexed RNA-FISH signals generated reproducible spatial expression maps in the developing chicken retina. C_LIO_LIResolution of relative boundaries for early patterning genes revealed nested expression domains marking the boundaries of the developing HAA. C_LIO_LIRNA-FISH experimental data using genes with axis-specific expression allowed the generation of DV and NT scores which enabled the localization of each cell in a scRNA-seq dataset onto a 2D topographical map of the developing retina. C_LIO_LI2D topographic maps derived from scRNA-seq data revealed novel HAA-enriched genes in the developing chicken retina. C_LIO_LICross-species comparisons showed conserved DV/NT programs but divergent fine-scale organization in retinal development. C_LIO_LICyp26c1 expression pattern was validated in human retina correlating with the location of the future HAA. C_LI

Matching journals

The top 5 journals account for 50% of the predicted probability mass.

50% of probability mass above

"Similar papers" are the closest papers from that journal in the model's embedding space. They show what the match is built on, but the ranking comes mostly from a classifier over the whole training set, not from these examples alone.