The term batch effects, often used in the biological community, describes a situation where subsets (batches) of the measurements significantly differ in distribution, due to irrelevant instrument-related factors. Batch effects introduce systematic error, which may cause statistical analysis to produce spurious results and/or obfuscate the signal of interest. Typically, the systematic effect of varying instrument conditions on the measurements depends on many unknown factors, whose impact on the difference between the observed and underlying true signal cannot be modeled.
An easy-to-use application for microarray, RNA-Seq and metabolomics analysis. For splicing sensitive platforms (RNA-Seq or Affymetrix Exon, Gene and Junction arrays), AltAnalyze will assess alternative exon (known and novel) expression along protein isoforms, domain composition and microRNA targeting. In addition to splicing-sensitive platforms, AltAnalyze provides comprehensive methods for the analysis of other data (RMA summarization, batch-effect removal, QC, statistics, annotation, clustering, network creation, lineage characterization, alternative exon visualization, gene-set enrichment and more).
Removes systematic batch effects. Batch Effect Removal is a method based on a residual neural network, trained to minimize the Maximum Mean Discrepancy (MMD) between the multivariate distributions of two replicates, measured in different batches. The software gives access to the datasets and models used to produce the results in the manuscript. The approach is general and can be applied to various data types.
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.
Proposes a set of methods for both RNA-seq and single-cell RNA-seq analysis. BBrowser supplies interactive visualizations as well as multiple analytics such as quality control, enrichment analysis, sub-clustering or differential expression. The platform includes a set of precomputed datasets encompassing research studies and sequencing samples or allows users to submit their own data.
Serves for feature selection on single-cell gene expression data. scGEApp allows users to detect genes playing a role in tissue structural remolding in idiopathic pulmonary fibrosis lungs. It makes two key contributions: (1) introducing a non-parametric, 3D spline-based feature selection method, and (2) defining a graphical user interface (GUI) for a number of commonly used methods in scRNA-seq data analysis.
Corrects batch effects, clusters cell types, imputes missing data caused by dropout events, and detects differentially expressed genes without requiring a preliminary normalization step. BUSseq consists of an interpretable hierarchical model that takes care of the count data nature, the overdispersion, the dropout events, and the cell-specific size factors of scRNA-seq data.
Assists in probabilistic representation and analysis of gene expression in single cells. scVI is a probabilistic approach to normalization and downstream analysis of scRNA-seq data. This method is based on a hierarchical Bayesian model with conditional distributions specified by deep neural networks. It permits to rely instead on stochastic optimization, sampling a fixed number of cells at each iteration.