ShinyButchR: Interactive NMF-based decomposition workflow of genome-scale datasets

Abstract

Non-negative Matrix Factorization (NMF) has been widely used for the analysis of genomic data to perform feature extraction and signature identification due to the interpretability of the decomposed signatures. However, running a basic NMF analysis requires the installation of multiple tools and dependencies, along with a steep learning curve and computing time. To mitigate such obstacles, we developed ShinyButchR, a novel R/Shiny application that provides a complete NMF-based analysis workflow, allowing the user to perform matrix decomposition using NMF, feature extraction, interactive visualization, relevant signature identification and association to biological and clinical variables. ShinyButchR builds upon the also novel R package ButchR, which provides new TensorFlow solvers for algorithms of the NMF family, functions for downstream analysis, a rational method to determine the optimal factorization rank and a novel feature selection strategy. ShinyButchR is publicly hosted at https://hdsu-bioquant.shinyapps.io/shinyButchR/, the source code is available at https://github.com/hdsu-bioquant/shinyButchR, and a Docker image at https://hub.docker.com/r/hdsu/shinybutchr. ButchR is freely available at https://github.com/wurst-theke/ButchR under the GPLv3 license, and a Docker image including test datasets is available at https://hub.docker.com/r/hdsu/butchr.

Genomic profiling of HIV-1 integration in microglia links viral insertions to TAD organization

“…HIV-1 integration association with genic and distal enhancers, including SE, was previously described in T cells (21, 29, 31). To investigate further the previously observed relation of genic enhancers with IS at nucleosomal resolution in microglia (Figure 2E) and to find other IS-associated genomic signatures on a larger genomic scale, we first employed a NMF-based approach recently developed (64). In this analysis, the genome was divided into 50 kb bins and each bin was characterized by its RNA-Seq, ChIP-Seq, and ATAC-Seq signal.…”

“…The NMF analysis was performed using the ButchR package (v1.0) over the summarized combined outputs of multiBigwigSummary over RNA-Seq, ChIP-Seq, and ATAC-Seq datasets (59, 64, 65). NMF computation was carried out with 10 4 iterations, 20 initializations, and rank factorization from 2 to 7.…”

Identification of Novel Subtypes in Lung Adenocarcinoma: Evidence from Gene Set Variation Analysis in Tumor and Adjacent Nontumor Samples

“…We used enrichment score (ES) as the activation degree of the gene sets. All LUAD patients were divided into 3 subconsensuses by NMF package [ 16 ] in R software based on the ES.…”

“…We used enrichment score (ES) as the activation degree of the gene sets. All LUAD patients were divided into 3 subconsensuses by NMF package [16] in R software based on the ES. [17] is a commonly used bioinformatic method to search for comprehensive information on large-scale genetic data.…”

Identification of Specific Cervical Cancer Subtypes and Prognostic Gene Sets in Tumor and Nontumor Tissues Based on GSVA Analysis

“…The nonnegative matrix factorization (NMF) was used to reduce the dimensions of complex data and provide a powerful assistance for clustering [ 17 ]. Based on the optimal number of clusters we found, we used k-means clustering algorithm to conduct the clustering analysis and draw a cluster diagram through NMF [ 18 ] and factoextra package. Survival analysis was applied to discover the difference among the divided subtypes, and the Kaplan-Meier plot was plotted at the same time.…”