ZipAEr: A compressive convolutional autoencoder for high-dimensional spatial omics data at subcellular resolution
Year: 2025
Type: preprint
Abstract: Recent advances in spatial transcriptomics have produced rich, high-throughput datasets, but their biological interpretation remains challenging due to analytical complexity. We present ZipAEr, a convolutional autoencoder tailored to extract informative latent features from spatial omics data. Unlike traditional methods that reduce data at the cell level, ZipAEr operates at the transcript level, preserving both subcellular and extracellular spatial context. Conventional autoencoders, built for images with three channels (red, green, blue), cannot handle spatial omics data with thousands of input channelsrepresenting genes and proteins. ZipAEr addresses this by reducing both spatial dimensions and channel count through its convolutional layers. It also introduces channel weighting in the loss function to ensure balanced representation of lowly expressed genes. ZipAEr effectively compresses spatial omics data by two to three orders of magnitude while preserving key spatial and molecular features. The resulting latent representation enables downstream analyses, such as classification and clustering, which would otherwise be computationally infeasible with raw data.
Source: Research Square
Language: English
Subfield: Molecular Biology
Domain: Life Sciences
Open Access status: green
Awards: Proteasome dysfunction as a driver of age-associated risk for Alzheimer's disease onset and progression, Development of an Innovative Vervet (Chlorocebus aethiops sabaeus) Model of Early Alzheimer's-like Neuropathology and Symptomatology, G178365870, Cell Senescence and Death in Neurodegenerative Diseases, DISCOVERY: Determinants of Incident Stroke Cognitive Outcomes and Vascular Effects on RecoverY, Investigating the role of transposable element dysregulation as a driver of neurotoxicity in tauopathy, G4766824899, Mechanisms of tau- and aging-induced neurological dysfunction: Focus on the nucleus, Evaluating microtubule binding as a potential imaging biomarker for Alzheimer's disease, G6553588769, Plasma Proteome and Risk of Alzheimer Dementia and Related Endophenotypes in the Framingham Study, Midwest Murine-Tissue Mapping Center (MM-TMC), Precursors of Stroke Incidence and Prognosis, High Resolution Profiling of Senescent Neurons and Their Microenvironments in Postmortem Human Brain Tissue Spanning Eight Decades of Life, South Texas Alzheimer's Disease Center, The pathogenic effects of senescent cells on their microenvironments and their role in human brain aging
OpenAlex ID: W4414295954