Hallinan, C. | Search bioRxiv

bioRxiv2024

Spatial mutual nearest neighbors for spatial transcriptomics data

Mutual nearest neighbors (MNN) is a widely used computational tool to perform batch correction for single-cell RNA-sequencing data. However, in applications such as spatial transcriptomics, it fails to take into account the 2D spatial information. Here, we present spatialMNN, an algorithm that integrates multiple spatial transcriptomic samples and identifies spatial domains. Our approach begins by building a k-Nearest Neighbors (kNN) graph based on the spatial coordinates, prunes noisy edges, and identifies niches to act as anchor points for each sample. Next, we construct a MNN graph across the samples to identify similar niches. Finally, the spatialMNN graph can be partitioned using existing algorithms, such as the Louvain algorithm to predict spatial domains across the tissue samples. We demonstrate the performance of spatialMNN using large datasets, including one with N=36 10x Genomics Visium samples. We also evaluate the computing performance of spatialMNN to other popular spatial clustering methods. Our software package is available at (https://github.com/Pixel-Dream/spatialMNN).

bioinformatics↗

bioRxiv2023

Heterogeneity-Preserving Discriminative Feature Selection for Subtype Discovery

The identification of disease-specific subtypes can provide valuable insights into disease progression and potential individualized therapies, important aspects of precision medicine given the complex nature of disease heterogeneity. The advent of high-throughput technologies has enabled the generation and analysis of various molecular data types, such as single-cell RNA-seq, proteomic, and imaging datasets, on a large scale. While these datasets offer opportunities for subtype discovery, they also pose challenges in finding subtype signatures due to their high dimensionality. Feature selection, a key step in the machine learning pipeline, involves selecting signatures that reduce feature size for more efficient downstream computational analysis. Although many existing methods focus on selecting features that differentiate known diseases or cell states, they often struggle to identify features that both preserve heterogeneity and reveal subtypes. To address this, we utilized deep metric learning-based feature embedding to explore the statistical properties of features crucial for preserving heterogeneity. Our analysis indicated that features with a notable difference in interquartile range (IQR) between classes hold important subtype information. Guided by this insight, we developed a statistical method called PHet (Preserving Heterogeneity), which employs iterative subsampling and differential analysis of IQR combined with Fishers method to identify a small set of features that preserve heterogeneity and enhance subtype clustering quality. Validation on public single-cell RNA-seq and microarray datasets demonstrated PHets ability to maintain sample heterogeneity while distinguishing known disease/cell states, with a tendency to outperform previous differential expression and outlier-based methods. Furthermore, an analysis of a single-cell RNA-seq dataset from mouse tracheal epithelial cells identified two distinct basal cell subtypes differentiating towards a luminal secretory phenotype using PHet-based features, demonstrating promising results in a real-data application. These results highlight PHets potential to enhance our understanding of disease mechanisms and cell differentiation, contributing significantly to the field of personalized medicine.

bioinformatics↗