Morgans, W. G. | Search bioRxiv

bioRxiv2023

CellPie: a fast spatial transcriptomics topic discovery method via joint factorization of gene expression and imaging data

Spatially resolved transcriptomics has enabled the study of expression of genes within tissues while retaining their spatial identity. Most spatial transcriptomics technologies generate a matched histopathological image as part of the standard pipeline, providing morphological information that can complement the transcriptomics data. Here we present CellPie, a fast, unsupervised factor discovery method, based on joint non-negative matrix factorisation of spatial RNA transcripts and histological image features.CellPie employs the accelerated hierarchical least squares method to significantly reduce the computational time, enabling efficient application to high-dimensional spatial transcriptomics datasets. We assessed CellPie on two different human cancer types and spatial resolutions, showing an improved performance against published factorisation methods. Additionally, we applied CellPie to a highly resolved Visium HD dataset, demonstrating its high computational efficiency compared to standard non-negative matrix factorisation and other existing methods. Availabilityhttps://github.com/ManchesterBioinference/CellPie

bioinformatics↗

bioRxiv2023

Scalable joint non-negative matrix factorisation for paired single cell gene expression and chromatin accessibility data

Single cell multi-modal technologies provide powerful means to simultaneously profile components of the gene regulatory path-ways of individual cells. These are now being employed to study gene regulatory mechanisms in a variety of biological systems. Tailored computational methods for integration and analysis of these data are much-needed with desirable properties in terms of efficiency -to cope with high dimensionality of the data, inter-pretability -for downstream biological discovery and hypothesis generation, and flexibility -to be able to easily incorporate future modalities. Existing methods cover some but not all of the desirable properties for effective integration of these data. Here we present a highly efficient method, intNMF, for representation and integration of single cell multi-modal data using joint non-negative matrix factorisation which can facilitate discovery of linked regulatory topics in each modality. We provide thorough benchmarking using large publicly available datasets against five popular existing methods. intNMF performs comparably against the current state-of-the-art, and provides advantages in terms of computational efficiency and interpretability of discovered regulatory topics in the original feature space. We illustrate this enhanced interpretability in providing insights into cell state changes associated with Alzheimers disease. int-NMF is available as a Python package with extensive documentation and use-cases at https://github.com/wmorgans/quick_intNMF

bioinformatics↗