iSSVD 0.1.5

Robust Integrative Biclustering for Multi-view Data

iSSVD - Intergrative Biclustering for Multi-view data with nested stability selection

Inputs:

X: A list contains multi-view data.

standr: If True each view will to be standardized. Default: False.

pointwise: If True a fast pointwise control method will be performed for stability selection. Default: True.

steps: Number of subsmaples used to perform stability selection. Default: 100.

size: Size of the subsamples used to perform stability selection. Default: 0.5.

vthr: The proportion to be explained by eigenvalues. Default: 0.7.

ssthr: Range of the threshold for stability selection. Default: [0.6, 0.8].

nbicluster: A user specified number of biclusters to be detected. Default: 10.

rows_nc: If True allows for negative correlation of rows over columns. Default: True.

cols_nc: If True allows for negative correlation of columns over rows. Default: True.

col_overlap: If True allows for columns overlaps among biclusters. Default: False.

row_overlap: If True allows for rows overlaps among biclusters. Default: False.

pceru: Per-comparrison wise error rate to control the number of falsely selected coefficients in the left singular vectors. Default: 0.1.

pcerv: Per-comparrison wise error rate to control the number of falsely selected coefficients in the right singular vector. Default: 0.1.

merr: Convergence threshold. Default: 1e-4.

iters: Maximal iteration for detecting each bicluster. Default: 100.

Outputs:

iSSVD returns the stable solutions of left singular vectors for each bicluster, and right singular vectors from each view for each bicluster.

N: Number of biclusters detected.

Info: Stability selection results of left and right singular vectors.

Sample_index: The indices of bicluster samples.

Variable_index: The indices of bicluster variables.

Iterations: The interations run for each bicluster.

Please check https://github.com/weijie25/iSSVD/blob/master/iSSVD/Guide.md for a simple guide.

Reference

Weijie Zhang and Sandra E. Safo. "Robust Integrative Biclustering for Multi-view Data." Bioinformatics submitted (2020).