Unraveling dynamically-encoded latent transcriptomic patterns in pancreatic cancer cells by topic modelling

Abstract

Building a comprehensive topic model has become an important research tool in single-cell genomics. With a topic model, we can decompose and ascertain distinctive cell topics shared across multiple cells, and the gene programs implicated by each topic can later serve as a predictive model in translational studies. Here, we present a Bayesian topic model that can uncover short-term RNA velocity patterns from a plethora of spliced and unspliced single-cell RNA-seq counts. We showed that modelling both types of RNA counts can improve robustness in statistical estimation and reveal new aspects of dynamic changes that can be missed in static analysis. We showcase that our modelling framework can be used to identify statistically-significant dynamic gene programs in pancreatic cancer data. Our results discovered that seven dynamic gene programs (topics) are highly correlated with cancer prognosis and generally enrich immune cell types and pathways.

Tree-structured topic modelling of single-cell gene expression data uncovers hierarchical relationships between immune cell types

“…We use the inference scheme as described in Kingma et al [9] implemented in Pytorch using torch.autograd [22,27]. Full variational inference and model training methods can be found in Zhang et al [28].…”

A highly scalable approach to topic modelling in single-cell data by approximate pseudobulk projection

“…In single-cell genomics, especially for sparse DNA accessibility data, embedding methods based on topic modelling (or latent semantic indexing), such as ArchR (Granja et al 2021) and cisTopic (Bravo González-Blas et al 2019), were ranked in the top lists of a recent benchmark study (Chen et al 2019). More recently, a topic model approach based on a deep variational autoencoder model, called embedded topic model, or ETM (Dieng et al 2020), has been successfully used in single-cell RNA-seq modelling (Zhao et al 2021b; Subedi and Park 2023; Zhang et al 2023).…”