Multi-Omic Integration via Sparse Singular Value Decomposition
Assign point color and shape aesthetics.
Adjust omic blocks for covariates effects.
Extracts (and merges) chunks of characters.
Multi-Omic integration via Sparse Singular value decomposition.
Creates a heatmap from the output of MOSS.
Returns features and subject selected by latent dimension.
Returns signatures of features by groups of subjects
Useful Venn diagrams to study the overlap between samples row names.
Mapping principal components onto a 2D map via tSNE.
Missing values imputation by the mean of each column.
Scale and normalize columns of a matrix.
Simple simulation of regulatory modules.
Sparse Singular Value Decomposition via Elastic Net.
'Solution path' for sparse Singular Value Decomposition via Elastic Ne...
'Solution path' for sparse Singular Value Decomposition via Elastic Ne...
t-Stochastic Neighbor Embedding to Clusters
High dimensionality, noise and heterogeneity among samples and features challenge the omic integration task. Here we present an omic integration method based on sparse singular value decomposition (SVD) to deal with these limitations, by: a. obtaining the main axes of variation of the combined omics, b. imposing sparsity constraints at both subjects (rows) and features (columns) levels using Elastic Net type of shrinkage, and c. allowing both linear and non-linear projections (via t-Stochastic Neighbor Embedding) of the omic data to detect clusters in very convoluted data (Gonzalez-Reymundez et. al, 2022) <doi:10.1093/bioinformatics/btac179>.