Poster Presentation 44th Lorne Genome Conference 2023

Disparities in spatially variable gene calling highlight the need for benchmarking spatial transcriptomics methods.  (#127)

Natalie C Charitakis 1 2 3 , Agus Salim 4 5 , Adam T Piers 1 3 , Kevin I Watt 1 3 6 , Enzo R Porrello 1 3 7 , David A Elliott 1 2 3 , Mirana Ramialison 1 3 8
  1. Stem Cell Biology, Murdoch Children's Research Institute, Parkville, Victoria, Australia
  2. Department of Paediatrics, University of Melbourne, Parkville, Victoria, Australia
  3. Novo Nordisk Foundation Center of Stem Cell Medicine, Parkville, Victoria, Australia
  4. Melbourne School of Population and Global Health, University of Melbourne, Carlton, Victoria, Australia
  5. School of Mathematics and Statistics, University of Melbourne, Parkville, Victoria, Australia
  6. Department of Diabetes, Monash University, Melbourne, Victoria, Australia
  7. Department of Anatomy and Physiology, University of Melbourne, Parkville, Victoria, Australia
  8. Australian Regenerative Medicine Institute, Monash University, Clayton, Victoria, Australia

Identifying spatially variable genes (SVGs) is a key step in the analysis of spatially resolved transcriptomics (SRT) data. SVGs provide biological insights by defining transcriptomic differences within tissues, which was previously unachievable using RNA-sequencing technologies. However, the increasing number of published tools designed to define SVG sets currently lack benchmarking methods to accurately assess performance. This study compares results of 5 purpose-built packages for SVG identification across 8 public and 3 simulated datasets and highlights discrepancies between results. Additional tools for generation of simulated data and development of benchmarking methods are required to improve methods for identifying SVGs.