| Title: | Data required for getting the gene set scores with NetActivity package |
|---|---|
| Description: | This package contains the weights from pre-trained shallow sparsely-connected autoencoders. This data is required for getting the gene set scores with NetActivity package. |
| Authors: | Carlos Ruiz-Arenas [aut, cre] |
| Maintainer: | Carlos Ruiz-Arenas <[email protected]> |
| License: | MIT + file LICENSE |
| Version: | 1.14.0 |
| Built: | 2026-10-10 11:16:03 UTC |
| Source: | https://github.com/bioc/NetActivityData |
We trained a shallow sparsely-connected autoencoder using GO biological processes and KEGG pathways as gene sets. The model was trained using all the GTEx samples (in total 7,729 samples for training). The matrix contains the weights obtained after training the model. Dimensions of the matrix correspond to the number of genes in the original matrix (8,758 genes) and the selected gene sets (1,518 gene sets). Although the weights sign is arbitrary, genes not included in a gene set have a weight of zero. Rownames correspond to gene set names (either GO term id or KEGG pathway id) while colnames correspond to the genes' ENSEMBL id.
A matrix with 1518 rows (gene sets) and 8758 columns (genes). Colnames and rownames include gene set and gene names.
We provide the annotation for the gene sets present in the GO_KEGG model trained with the GTEx samples. This object contains the GO terms and KEGG pathways names as well as the genes weights.
A data.frame with 1518 rows corresponding to (gene sets).
Term Gene set full name
GeneSet Gene set ID. This ID matches the gene set id in the weights' matrix.
Weights List with gene weights used to compute the gene set score. Gene names correspond to ENSEMBL ids.
Weights_SYMBOL List with gene weights used to compute the gene set score. Gene names correspond to SYMBOL ids.
NetActivityData contains the weights from pre-trained shallow sparsely-connected autoencoders. This data is required for getting the gene set scores with NetActivity package.
We trained a shallow sparsely-connected autoencoder using GO biological processes and KEGG pathways as gene sets. The model was trained using TCGA samples from multiple tumors (in total 8,874 samples for training). The matrix contains the weights obtained after training the model. Dimensions of the matrix correspond to the number of genes in the original matrix (8,758 genes) and the selected gene sets (1,518 gene sets). Although the weights sign is arbitrary, genes not included in a gene set have a weight of zero. Rownames correspond to gene set names (either GO term id or KEGG pathway id) while colnames correspond to the genes' ENSEMBL id.
A matrix with 1518 rows (gene sets) and 8758 columns (genes). Colnames and rownames include gene set and gene names.
We provide the annotation for the gene sets present in the GO_KEGG model trained with the TCGA samples. This object contains the GO terms and KEGG pathways names as well as the genes weights.
A data.frame with 1518 rows corresponding to (gene sets).
Term Gene set full name
GeneSet Gene set ID. This ID matches the gene set id in the weights' matrix.
Weights List with gene weights used to compute the gene set score. Gene names correspond to ENSEMBL ids.
Weights_SYMBOL List with gene weights used to compute the gene set score. Gene names correspond to SYMBOL ids.