Infers missing values. LSImpute works with several steps: (i) it selects pairs of cells with similarity level, (ii) it clusters the cells previously selected, (iii) for each of identified clusters, it replaces zero values for each gene with values imputed based on the expression levels of gene in all the cells within the cluster, and (iv) the selected cells have imputed values and the clusters they form are collapsed into their respective centroids.
Allows studying of spatial patterning of gene expression at the single-cell level. Seurat is an R package that enables quality control (QC), analysis, and exploration of single cell RNA-seq data. The software includes three computational methods: (1) unsupervised clustering and discovery of cell types and states, (2) spatial reconstruction of single cell data, and (3) integrated analysis of single cell RNA-seq across conditions, technologies, and species. It can also localize rare subpopulations, and map both spatially restricted and scattered groups.
An ultrafast algorithm which uses a novel yet very simple "implicit imputation" approach to alleviate the impact of dropouts in single cell RNA-seq (scRNA-seq) data in a principled manner. Using a range of simulated and real data, we have shown that CIDR outperforms the state-of-the-art methods, namely t-SNE, ZIFA and RaceID, by at least 50% in terms of clustering accuracy, and typically completes within seconds for processing a dataset of hundreds of cells. We believe that single-cell mRNA sequencing in combination with the RaceID algorithm is a powerful tool to unravel heterogeneity of rare cell types in both healthy and diseased organs.
Offers a method for handling and extracting structure from single-cell RNA-sequencing and CyTOF data. SAUCIE is a standalone software that provides a deep learning approach developed for the analysis of single-cell data from a cohort of patients. The application is based on different layers able to performs several tasks such as data imputation, clustering, batch correction or visualization. The approach is based on the autoencoder neural network framework for unsupervised learning.
Provides a method for the iterative normalization and cluster of single-cell gene RNA-seq expression data. BISCUIT eases clustering of cells based on similar gene expression after correcting technical variation. The software uses a Bayesian model and employs a model driven by covariance structures for the normalization and the input of data. These functionalities are appropriate to work on tumor heterogeneity and other primary tissue to understand novel cell types.
Address the dropout events prevalent in scRNAseq data. scImpute is an imputation method that determines which values are affected by dropout events in data and performs imputation only on dropout entries. This method learns each gene’s dropout probability in each cell by fitting a mixture model. Next, it imputes the dropout values in a cell by borrowing information of the same gene in other similar cells.
Imputes dropout values in single-cell RNA sequencing (scRNA-seq) data. DeepImpute is a package exploiting deep neural networks and models based on sub-networks derived from subsets generated from split genes. This program proposes a method enabling the use of a model which can be trained with a subset of data to increase the rapidity of its analysis. It was tested on four different datasets.