WebAtlas Visualiser of Multiplexed Error-Robust Fluorescence In Situ Hybridisation (MERFISH) dataset in wheat inflorescence tissue
This spatial transcriptomics resource features MERFISH data from a 200-gene panel targeting wheat inflorescence tissues of hexaploid bread wheat (cv 'Paragon'). The project is a collaboration between Vizgen, the Earlham Institute, and the John Innes Centre.
Probes for 200 genes were designed and applied using the VizGen MERSCOPE platform on wheat inflorescence tissues at four developmental stages, defined according to the Waddington scale (W2.5 Late Double Ridge; W3.25 Lemma Primordium; W4 Terminal Spikelet and; W5 Carpel Extending Round; see here for details on staging). We utilized the VizGen post-processing tool¹ and Cellpose (v2.2.3)² for image segmentation and transcript assignment. For data analysis, we applied the Scanpy³, Squidpy⁴, and Scanorama⁵ packages to perform Leiden clustering and marker gene identification. In total, we defined 18 expression domains, which were assigned to specific tissue types based on their spatial position and anatomical characteristics. The location of expression domains can be found under 'Domain' tabs in the visualiser tool. Gene expression levels in the tool are summarised by total counts per cell.
To facilitate the use of this resource we include:
- List of 200 genes used for probe design (including homoeologs and annotations) (link to csv)
- Normalised expression values of 200 genes across the expression domains in each of the four samples (link to csv)
This work was led by Katie Long and Cristobal Uauy from the John Innes Centre, Ashleigh Lister and Iain Macaulay from the Earlham Institute, and Carole Chedid and James Hargreaves from Vizgen. We thank Chen Ji and Philippa Borrill (JIC), Xuemei Liu and Jun Xiao (IGDB-CAS), Max Jones and Yuanrong Pei (JIC) and additional members of the Uauy lab for their suggestions in defining the gene set. Special thanks to Rob Ellis, Martin Vickers and Burkhard Steuernagel from JIC Informatics.
The implementation of this instance of WebAtlas was led by Rob Ellis, in close coordination with Katie Long.
We also thank the Haniffa Lab, Bayraktar Lab, and Teichmann Lab for our use of the open-source WebAtlas pipeline.