ProtoCell4P: an explainable prototype-based neural network for patient classification using single-cell RNA-seq

Abstract

Motivation The rapid advance in single-cell RNA sequencing (scRNA-seq) technology over the past decade has provided a rich resource of gene expression profiles of single cells measured on patients, facilitating the study of many biological questions at the single-cell level. One intriguing research is to study the single cells which play critical roles in the phenotypes of patients, which has the potential to identify those cells and genes driving the disease phenotypes. To this end, deep learning models are expected to well encode the single-cell information and achieve precise prediction of patients’ phenotypes using scRNA-seq data. However, we are facing critical challenges in designing deep learning models for classifying patient samples due to (1) the samples collected in the same dataset contain a variable number of cells — some samples might only have hundreds of cells sequenced while others could have thousands of cells, and (2) the number of samples available is typically small and the expression profile of each cell is noisy and extremely high-dimensional. Moreover, the black-box nature of existing deep learning models makes it difficult for the researchers to interpret the models and extract useful knowledge from them. Results We propose a prototype-based and cell-informed model for patient phenotype classification, termed ProtoCell4P, that can alleviate problems of the sample scarcity and the diverse number of cells by leveraging the cell knowledge with representatives of cells (called prototypes), and precisely classify the patients by adaptively incorporating information from different cells. Moreover, this classification process can be explicitly interpreted by identifying the key cells for decision making and by further summarizing the knowledge of cell types to unravel the biological nature of the classification. Our approach is explainable at the single-cell resolution which can identify the key cells in each patient’s classification. The experimental results demonstrate that our proposed method can effectively deal with patient classifications using single-cell data and outperforms the existing approaches. Furthermore, our approach is able to uncover the association between cell types and biological classes of interest from a data-driven perspective. Availability https://github.com/Teddy-XiongGZ/ProtoCell4P

GraSP Gene Targets to Hierarchically Infer Sub-Classes with CuttleNet

“…We finally benchmarked our networks against CellTICS, the most cutting-edge published hierarchical NN model for simultaneous cell class and subclass classification (Yin & Chen, 2023). Appreciating the extensive benchmarking of CellTICS against other leading methods, and its superior performance, we deemed it the best reference for evaluating our models.…”

“…Here we bridge this gap by contributing: 1) A novel neural network (NN) training strategy for efficient dimensionality reduction. As well as, 2) CuttleNet, a hierarchical NN architecture inspired by coarse-to-fine biological perception, achieving state-of-the-art performance scores benchmarked against the best available supervised classifier for this problem (Yin & Chen, 2023); outperforming it on our specific application.…”

DeepGSEA: Explainable Deep Gene Set Enrichment Analysis for Single-cell Transcriptomic Data

“…They interpreted the model by looking at its learned weights on different prototypes for the classification task. For scRNA-seq data, we proposed ProtoCell4P [52] which leverages the interpretability of prototype-based models to perform intelligible patient classification. The model learns cell type-informed prototypes of cells and classifies each patient by summarizing prototype information within each cell.…”

“…By measuring the expression profiles of individual cells, researchers identified substantial cellular heterogeneity in gene expression that was not found in bulk RNA-seq data [28]. To deal with cellular heterogeneity in single-cell analysis, a number of deep learning (DL) methods have been proposed to model the complex gene expression distributions and in turn, handle canonical processing of scRNA-seq data, such as gene expression normalization [6], batch correction [27], clustering [13, 41], cell type identification [7, 18], and patient classification [15, 52].…”

“…With the rapid development of explainable artificial intelligence, a number of ante-hoc explainable methods have been proposed to build DL models that are both powerful and interpretable [17, 19, 40], among which prototype-based architecture is a promising branch that can provide case-based reasoning for trained models [7, 26, 52]. In this paper, we provide a solution to leverage the expressiveness of DNNs for GSE analysis while preserving interpretability for the prediction.…”