TY - GEN
T1 - Mapping the Topography of Spatial Gene Expression with Interpretable Deep Learning
AU - Chitra, Uthsav
AU - Arnold, Brian J.
AU - Sarkar, Hirak
AU - Ma, Cong
AU - Lopez-Darwin, Sereno
AU - Sanno, Kohei
AU - Raphael, Benjamin J.
N1 - Publisher Copyright:
© The Author(s), under exclusive license to Springer Nature Switzerland AG 2024.
PY - 2024
Y1 - 2024
N2 - Spatially resolved transcriptomics technologies provide high-throughput measurements of gene expression in a tissue slice, but the sparsity of this data complicates the analysis of spatial gene expression patterns. We address this issue by deriving a topographic map of a tissue slice—analogous to a map of elevation in a landscape—using a novel quantity called the isodepth. Contours of constant isodepth enclose spatial domains with distinct cell type composition, while gradients of the isodepth indicate spatial directions of maximum change in gene expression. We develop GASTON, an unsupervised and interpretable deep learning algorithm that simultaneously learns the isodepth, spatial gene expression gradients, and piecewise linear functions of the isodepth. GASTON models both continuous gradients and discontinuous spatial variation in the expression of individual genes. We show that GASTON accurately identifies spatial domains and marker genes in multiple SRT datasets.
AB - Spatially resolved transcriptomics technologies provide high-throughput measurements of gene expression in a tissue slice, but the sparsity of this data complicates the analysis of spatial gene expression patterns. We address this issue by deriving a topographic map of a tissue slice—analogous to a map of elevation in a landscape—using a novel quantity called the isodepth. Contours of constant isodepth enclose spatial domains with distinct cell type composition, while gradients of the isodepth indicate spatial directions of maximum change in gene expression. We develop GASTON, an unsupervised and interpretable deep learning algorithm that simultaneously learns the isodepth, spatial gene expression gradients, and piecewise linear functions of the isodepth. GASTON models both continuous gradients and discontinuous spatial variation in the expression of individual genes. We show that GASTON accurately identifies spatial domains and marker genes in multiple SRT datasets.
KW - Spatial transcriptomics
KW - deep learning
KW - expression gradients
KW - gene expression topography
UR - http://www.scopus.com/inward/record.url?scp=85194226399&partnerID=8YFLogxK
UR - http://www.scopus.com/inward/citedby.url?scp=85194226399&partnerID=8YFLogxK
U2 - 10.1007/978-1-0716-3989-4_33
DO - 10.1007/978-1-0716-3989-4_33
M3 - Conference contribution
AN - SCOPUS:85194226399
SN - 9781071639887
T3 - Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)
SP - 368
EP - 371
BT - Research in Computational Molecular Biology - 28th Annual International Conference, RECOMB 2024, Proceedings
A2 - Ma, Jian
PB - Springer Science and Business Media Deutschland GmbH
T2 - 28th International Conference on Research in Computational Molecular Biology, RECOMB 2024
Y2 - 29 April 2024 through 2 May 2024
ER -