Software

Hypergraph Analysis Toolbox (HAT)

HAT is a a software package encompassing a range of techniques to identify, investigate, and visualize multi-way interactions in biological data. As a general purpose tool, the algorithms implemented in HAT address hypergraph construction, visualization, and the analysis of structural and dynamic properties. HAT is the first software to utilize tensor algebra for hypergraph analysis, and it contains recently developed methods for hypergraph similarity measures. [Code]

CHEbyshev Spectral HyperlInk pREdictor (CHESHIRE)

CHESHIRE is a deep learning-based method designed for hyperlink prediction on hypergraphs, specifically tailored to forecast missing reactions within genome-scale metabolic models (GEMs) based solely on metabolic network topology. CHESHIRE outperforms other topology-based methods in predicting artificially removed reactions over 926 high- and intermediate-quality GEMs and is able to improve the phenotypic predictions of 49 draft GEMs for fermentation products and amino acids secretions. CHESHIRE is a powerful tool for GEM curation to reveal unknown links between reactions and observed metabolic phenotypes. [Code]

GRAph Convolutional nEtwork feature Selector (GRACES)

GRACES is a deep learning-based method designed for feature selection in high-dimensional and low-sample size (HDLSS) data. GRACES exploits latent relations between samples with various overfitting-reducing techniques to iteratively find a set of optimal features which gives rise to the greatest decreases in the optimization loss. GRACES significantly outperforms other HDLSS feature selection methods on both synthetic and real-world datasets. [Code]

4DNvestigator

4DNvestigator is a user-friendly network-based toolbox for the analysis of time series genome-wide genome structure (Hi-C) and gene expression (RNA-seq) data by drawing on network theory, information theory, and multivariate statistics. 4DNvestigator encompasses various methods to quantify network entropy, tensor entropy, and statistically significant changes in time series Hi-C data at different genomic scales, which are important for producing rigorous quantitative results in 4DN research. [Code]