Super-resolved spatial transcriptomics by deep data fusion

Abstract

In situ RNA capturing has made it possible to record histology and spatial gene expression from the same tissue section. Here, we introduce a method that combines data from both modalities to infer super-resolved full-transcriptome expression maps. Our method unravels transcriptional heterogeneity in micrometer-scale anatomical features and enables image-based in silico spatial transcriptomics without hybridization or sequencing.

Efficient prediction of a spatial transcriptomics profile better characterizes breast cancer tissue sections without costly experimentation

“…We performed XFuse 5 , which is the method to generate super-resolved transcriptomics images using generative models (Supplementary Fig. S11 ).…”

“…Spatial transcriptomics with in situ capturing is an emerging technology that maps gene-expression profiles with corresponding spatial information in a tissue section 1 – 4 . A highly resolved spatial-transcriptome profile is invaluable for revealing biological functions and molecular mechanisms 5 . Recently, many histological transcriptome profiles, measured by in situ capturing platforms (numerous spots with barcoded oligonucleotides on a chip), were reported in the field of oncology 6 , 7 .…”

Deep topographic proteomics of a human brain tumour

“…This bulk property means that the data still suffers from missing values which can confound spatial pathway level analysis. Future approaches are bound to use systematic spatial proteomic analysis, possibly compromising spatial resolution but incorporating an element of machine learning to use orthogonal higher resolution omics and imaging data to infer protein abundance towards individual cell resolution, as can be done on spatial transcriptomics data [58][59][60][61] . In addition, with the detection of LCM-based and cell-type resolved deep proteomes, these data will be highly complementary to current imaging technologies and increase the understanding of spatially resolved biological and pathological processes at the molecular level 29,32,57 .…”

Emerging artificial intelligence applications in Spatial Transcriptomics analysis

“…XFuse [81] uses a Bayesian deep generative model to enhance the resolution and impute spatial gene expression with histological images. XFuse assumes the gene expression and histological image share an underlying latent state.…”