swCAM: estimation of subtype-specific expressions in individual samples with unsupervised sample-wise deconvolution

Abstract

Motivation Complex biological tissues are often a heterogeneous mixture of several molecularly distinct cell subtypes. Both subtype compositions and subtype-specific expressions can vary across biological conditions. Computational deconvolution aims to dissect patterns of bulk tissue data into subtype compositions and subtype-specific expressions. Existing deconvolution methods can only estimate averaged subtype-specific expressions in a population, while many downstream analyses such as inferring co-expression networks in particular subtypes require subtype expression estimates in individual samples. However, individual-level deconvolution is a mathematically underdetermined problem because there are more variables than observations. Results We report a sample-wise Convex Analysis of Mixtures (swCAM) method that can estimate subtype proportions and subtype-specific expressions in individual samples from bulk tissue transcriptomes. We extend our previous CAM framework to include a new term accounting for between-sample variations and formulate swCAM as a nuclear-norm and ℓ2,1-norm regularized matrix factorization problem. We determine hyperparameter values using cross-validation with random entry exclusion and obtain a swCAM solution using an efficient alternating direction method of multipliers. Experimental results on realistic simulation data show that swCAM can accurately estimate subtype-specific expressions in individual samples and successfully extract co-expression networks in particular subtypes that are otherwise unobtainable using bulk data. In two real-world applications, swCAM analysis of bulk RNASeq data from brain tissue of cases and controls with bipolar disorder or Alzheimer’s disease identified significant changes in cell proportion, expression pattern and co-expression module in patient neurons. Comparative evaluation of swCAM versus peer methods is also provided. Availability The R Scripts of swCAM are freely available at https://github.com/Lululuella/swCAM. A user’s guide and a vignette are provided. Supplementary information Supplementary data are available at Bioinformatics online.

Penalised regression improves imputation of cell-type specific expression using RNA-seq data from mixed cell populations compared to domain-specific methods

“…We provide a real data comparison of sample-level cell type specific expression imputation, including off-the-shelf machine learning methods, multivariate LASSO and RIDGE, as comparators. Correlation has been used to evaluate the accuracy of predicted cell-type expression, and good correlations per subject have been reported 13,20,23 , consistent with our observations. However, we also found high correlations in between-subject comparisons (Supplementary Figure 8), which presumably reflects that cell type explains the greatest proportion of variability in gene expression.…”

Benchmarking and integration of methods for deconvoluting spatial transcriptomic data

“…Another limitation of the current existing ST deconvolution methods is that most of them only infer cell type proportions but do not estimate cell-type-specific (CTS) gene expression at each spot, which is equally important for ST data analysis. For bulk RNA-seq data, many methods have been developed for CTS genes expression inference, such as TCA ( Rahmani et al , 2019 ), CIBERSORTx ( Newman et al , 2019 ), bMIND ( Wang et al , 2021 ) and swCAM ( Chen et al , 2022 ). However, for the ST data, only RCTD can infer the CTS gene expression at each spot.…”

Evaluating performance and applications of sample-wise cell deconvolution methods on human brain transcriptomic data

“…Cell-type gene expressions can be deconvoluted from bulk-tissue expression data. Methods have been developed to estimate cell-type expressions for each sample, such as bMIND( 28 ), swCAM( 29 ), and TCA( 30 ). We call this sample-wise deconvolution of gene expression.…”