SISUA: Semi-Supervised Generative Autoencoder for Single Cell Data

Abstract

Single-cell transcriptomics offers a tool to study the diversity of cell phenotypes through snapshots of the abundance of mRNA in individual cells. Often there is additional information available besides the single cell gene expression counts, such as bulk transcriptome data from the same tissue, or quantification of surface protein levels from the same cells. In this study, we propose models based on the Bayesian generative approach, where protein quantification available as CITE-seq counts from the same cells are used to constrain the learning process, thus forming a semi-supervised model. The generative model is based on the deep variational autoencoder (VAE) neural network architecture.Keywords semi-supervised · single-cell · RNA sequencing · deep learning · Bayesian inference 1 IntroductionSingle-cell RNA sequencing (scRNA-seq) [1,2,3] is a powerful tool to analyze cell states based on their gene expression profile with high resolution. RNA sequencing at single-cell level facilitates uncovering heterogeneous gene expression patterns in seemingly homogeneous cell populations. However, the current methods for gene expression profiling at single cell resolution are prone to experimental errors, in particular, inefficient capture of mRNAs [2]. This capture inefficiency results into a general underestimation of the counts (dropout effect). This represents a problem as the current computational approaches for analyzing single-cell data rely on the mRNA counts for clustering and downstream analysis.Generally, the solution to the dropout problem has been posed as an imputation task, where missing counts are filled with estimated counts. Different methods have been proposed for this task, such as non-negative regression [4] or graph-based methods [5]. Another option is to model the dropout effect using the zero-inflated (ZI) model [6], where a two-component mixture distribution is constructed, such that the first component models the dropout effect and the second component the observed counts. The effect of overdispersion is strongly presented in the scRNA-seq counts, the negative binomial (NB) distribution is seen as an appropriate fit to the observed data [7]. Shallow imputation models that are based on zero-inflated negative binomial (ZINB) or zero-inflated log-normal models have been applied to single-cell data [8,9]. However, these models hypothesize a linear relation between the latent space and the model parameters, which is quite a strong assumption [10]. To overcome the limitations of the linear models, deep neural network architectures have been proposed to resolve missing data (dropouts) [11]. However, discerning technical A PREPRINT -MAY 8, 2019 variation from biological signal solely based on scRNA-seq data is challenging, and assumes that a large number of similar cells are measured.Accurate imputation strategies are important for downstream analysis, including identification of cell type marker genes, characterization of functional state [12], or the analysis of transcriptome dynamics along differentia...

BaySiCle: A Bayesian Inference joint kNN method for imputation of single-cell RNA-sequencing data making use of local effect

“…Sample RNA-Sequencing data was provided with courtesy by Dr Ville Hautamäki who is the author of paper (Trung Ngo Trong et al, 2020), in which Bayesian inference method has also been used to get the latent values. However, in that paper, since the cells coming from a common group are not taken into account, they make use of another distribution (Kingma and Welling, 2014) to get the posterior probability.…”

“…However, in that paper, since the cells coming from a common group are not taken into account, they make use of another distribution (Kingma and Welling, 2014) to get the posterior probability. Our method of Bayesian inference to get the posterior probability for the values of gene expression which needs imputation relies a lot on the batch or group effect, and so the formula is designed accordingly unlike in Trung Ngo Trong et al, 2020, which did not take batch effect into account for their advantage. The sample data is also provided in the GitHub link specified to download the Python script.…”

A Joint Model of RNA Expression and Surface Protein Abundance in Single Cells

“…Previous work derives a linear cutoff on the number of counts for each protein based on spiked-in cells that do not express the proteins specifically recognized by the barcoded antibodies [3]. Others fit mixture models to each protein, which assumes that all cells are subject to the same background distribution [11,12]. Our approach, which obviates the need for negative control cells or the assumption of a constant background distribution, models each y nt |z n , µ nt as a negative binomial mixture, where the Bernoulli parameter π nt |z n can be interpreted as the probability that a cell's protein count came from the background.…”

“…As a result, the distribution of protein counts across cells is often bimodal with a background and foreground component. A natural way to preprocess the protein counts is to fit a mixture model to each protein globally and replace a count with its probability of being generated from the larger component [11,12]. However, there is no basis for assuming global bimodality of protein counts since a dataset typically comprises heterogeneous populations of cells each with distinct surface proteomes.…”

Semi-supervised single-cell cross-modality translation using Polarbear

“…In the first phase, we train two deep variational autoencoder (VAE) neural networks that learn, in an unsupervised fashion, to reduce each given type of data to a latent representation (the encoder) and then expand that representation to recover the original data (the decoder). (V)AEs have already been successfully applied to scRNA-seq and scATAC-seq data, primarily for the purpose of de-noising [11][12][13][14][15][16][17]. Polarbear trains one VAE for each type of data, while taking into consideration sequencing depth and batch factors [15,16].…”