Spatial transcriptomics is a method for assigning cell types (identified by the mRNA readouts) to their locations in the histological sections. It comprises an important part of spatial biology. Recent work demonstrated that the subcellular localization of mRNA molecules, for example, in the nucleus can also be studied.[1]
The first widely-adopted method, described by Ståhl et al., in a landmark 2016 paper in Science[2] implies positioning individual tissue samples on the glass slides or arrays including "spots" of spatially barcoded oligo(dT) tails to capture the poly-adenylated mRNAs.[2] Besides oligo(dT) tail and spatial barcode, which indicates the x and y position on the arrayed slide, the probe contains an amplification and sequencing handle, and unique molecular identifier.[2] Commonly, histological samples are cut using cryotome, then fixed, stained, and put on the microarrays.[2] After that, it undergoes enzymatic permeabilization, so that RNA molecules can diffuse down to the slide, with further hybridization of polyadenylated mRNA molecules to the oligo (dT) probes.[2] Reverse transcription is then carried out in situ.[2] As a result, spatially marked complementary DNA (cDNA) is synthesized, providing information about gene expression in the exact location of the sample.[2] From the cDNA, libraries are synthesised for short-read sequencing. In summary, the first widely-adopted spatial transcriptomics protocol combines paralleled sequencing and staining of the same sample.[2] In the downstream analysis, bioinformatic tools allow to overlap the tissue image with the gene expression. The output is a map of the transcriptome capture the gene expression of individual cells within a tissue section. It is important to mention that the first generation of the arrayed slides comprised about 1,000 spots of the 100-μm diameter, limiting resolution to ~10-40 cells per spot.[2]
Spatial transcriptomics includes methods that can be divided into two modalities, those based in next-generation sequencing for gene detection, and those based in imaging.[3] Some common approaches to resolve spatial distribution of transcripts are microdisection techniques, Fluorescent in situ hybridization methods, in situ sequencing, in situ capture protocols and in silico approaches.[4]