RNA Detection

Chapter 7

Spatial Transcriptomics: Constructing a Single-Cell

Resolution Transcriptome-Wide Expression Atlas

Kaia Achim, Hernando Martı ́nez Vergara, and Jean-Baptiste Pettit

Abstract

The method described here aims at the construction of a single-cell resolution gene expression atlas for an
animal or tissue, combining in situ hybridization (ISH) and single-cell mRNA-sequencing (scRNAseq).
A high resolution and medium-coverage gene expression atlas of an animal or tissue of interest can be
obtained by performing a series of ISH experiments, followed by a process of image registration and gene
expression averaging. Using the overlapping fraction of the genes, concomitantly obtained scRNAseq data
can be fitted into the spatial context of the gene expression atlas, complementing the coverage by genes.

Key wordsSpatial transcriptomics, Image registration, Single-cell mRNA-seq, Gene expression

1 Introduction

The molecular composition of cells defines their development, identity

and function. mRNA expression isone of the components of this

molecular composition. Recent advances in RNA expression studies

provideexcellent toolsfor the assessment ofthispart ofcellular identity.

The method described here is an integrative approach that com-

bines two cutting-edge high-throughput methods: universally appli-

cable single-cell mRNA sequencing techniques, and the generation

of ISH-based gene expression atlases. The integration of these two

technologies combines their respective advantages, that is, transcrip-

tomics and spatial profiling, allowing the generation of single-cell

resolution transcriptome-wide full-body expression atlases.

For performing the ISH and scRNA-seq, researcher is free to

use their favorite established protocol, such as ones described in this

book. In this chapter, we summarize, first, the requirements for the

acquisition and processing of the imaged ISH data to generate an

expression atlas; second, the methods for single-cell dissociation

and for obtaining an initial quality-control of the raw scRNAseq

read count data; and third, the computational pipeline for the

integration of these data (spatial mapping).

Imre Gaspar (ed.),RNA Detection: Methods and Protocols, Methods in Molecular Biology,
vol. 1649, DOI 10.1007/978-1-4939-7213-5_7,©Springer Science+Business Media LLC 2018

111