Package: SmCCNet 2.0.7

Abhinav Pundir

SmCCNet: Sparse Multiple Canonical Correlation Network Analysis Tool ('SmCCNet')

A canonical correlation based framework ('SmCCNet') designed for the construction of phenotype-specific multi-omics networks. This framework adeptly integrates single or multiple omics data types along with a quantitative or binary phenotype of interest. It offers a streamlined setup process that can be tailored manually or configured automatically, ensuring a flexible and user-friendly experience. Methods are described in Shi et al. (2019) "Unsupervised discovery of phenotype-specific multi-omics networks" <doi:10.1093/bioinformatics/btz226>.

Authors:Abhinav Pundir [cre], Weixuan Liu [aut], Yonghua Zhuang [aut], W. Jenny Shi [aut], Thao Vu [aut], Iain Konigsberg [aut], Katherine Pratte [aut], Laura Saba [aut], Katerina Kechris [aut]

SmCCNet_2.0.7.tar.gz
SmCCNet_2.0.7.zip(r-4.7-any)SmCCNet_2.0.7.zip(r-4.6-any)SmCCNet_2.0.7.zip(r-4.5-any)
SmCCNet_2.0.7.tgz(r-4.6-any)SmCCNet_2.0.7.tgz(r-4.5-any)
SmCCNet_2.0.7.tar.gz(r-4.7-any)SmCCNet_2.0.7.tar.gz(r-4.6-any)
SmCCNet_2.0.7.tgz(r-4.6-emscripten)
manual.pdf |manual.html✨
DESCRIPTION |NEWS
card.svg |card.png
SmCCNet/json (API)

# Install 'SmCCNet' in R:
install.packages('SmCCNet', repos = c('https://kechrislab.r-universe.dev', 'https://cloud.r-project.org'))

Bug tracker:https://github.com/kechrislab/smccnet/issues

Pkgdown/docs site:https://kechrislab.github.io

Datasets:
  • X1 - A synthetic mRNA expression dataset.
  • X2 - A synthetic miRNA expression dataset.
  • Y - A synthetic phenotype dataset.

On CRAN:

Conda:

network

6.68 score 32 stars 33 scripts 386 downloads 1 mentions 16 exports 45 dependencies

Last updated from:b863af0209. Checks:9 OK. Indexed: yes.

TargetResultTimeFilesSyslog
linux-develOK171
source / vignettesOK297
linux-releaseOK146
macos-releaseOK172
macos-oldrelOK213
windows-develOK105
windows-releaseOK89
windows-oldrelOK80
wasm-releaseOK146

Exports:%>%aggregateCVSingleclassifierEvaldataPreprocessfastAutoSmCCNetgetAbargetCanCorMultigetCanWeightsMultigetOmicsModulesgetRobustWeightsMultigetRobustWeightsMultiBinarygetRobustWeightsSinglegetRobustWeightsSingleBinarynetworkPruningscalingFactorInputsummarizeNetSHy

Dependencies:clicodetoolscpp11data.tabledigestEnvStatsfarverfurrrfutureggplot2globalsgluegtableigraphisobandjsonlitelabelinglatticelifecyclelistenvmagrittrMASSMatrixnnetnortestparallellypbapplypkgconfigplspracmapROCpurrrR6RColorBrewerRcpprlangrlistS7scalessplsvctrsviridisLitewithrXMLyaml

Automated SmCCNet
Function Arguments and Tuning Parameters | Examples
Rendered fromSmCCNet_Vignette_AutoSmCCNet.Rmdusingknitr::rmarkdown

Last update: 2026-05-08
Started: 2023-10-03

Reconstructing Phenotype-Specific Multi-Omics Networks with SmCCNet
SmCCNet overview | Workflow | SmCCNet package | SmCCNet workflow with a synthetic dataset | Synthetic dataset | Step I. Preprocessing | Step II: Determine optimal sparsity penalties through cross-validation (optional) | Create test and training data sets. | Set Scaling Factors | Run K-fold CV | Extract penalty pair with the smallest total prediction error | Step III: Run SmCCA with pre-selected penalty term | Step IV: Obtain multi-omics modules through network clustering | Step V: Obtain network summarization score and pruned subnetworks | Results | Step VI: Visualize network module | Methods for Optimal Scaling Factors Selection | Method 1: Use Prompt to Define Scaling Factors | Method 2: Use Automated SmCCNet to Select Scaling Factors Based on Pairwise Correlation. | Method 3: Use Cross-Validation to Select Scaling Factors | Cross validation with each set of scaling factors | Select optimal scaling factors with associated penalty parameters | Multi-Omics SmCCNet with Binary Phenotype | Extract penalty terms with the highest testing evaluation score | Step III: Run SPLSDA with pre-selected penalty terms | Acknowledgments | Session info | References
Rendered fromSmCCNet_Vignette_MultiOmics.Rmdusingknitr::rmarkdown

Last update: 2026-05-08
Started: 2023-10-03

Reconstructing Phenotype-Specific Single-Omics Networks with SmCCNet
SmCCNet package | SmCCNet single-omics workflow with a synthetic dataset | Step I. Preprocessing | Step II: Determine optimal sparsity penalties through cross-validation (optional) | Create test and training data sets. | Run K-fold Cross-Validation | Extract penalty term with the smallest total prediction error | Step III: Run SmCCA with pre-selected penalty term | Step IV: Obtain single-omics modules through network clustering | Step V: Obtain network summarization score and pruned subnetworks | Results | Step VI: Visualize network module | Single-omics SmCCNet for Binary Phenotype | Import synthetic dataset with binary phenotype | Step II: Determine optimal sparsity penalty through cross-validation (optional) | Step III: Run SPLSDA with pre-selected penalty term | Session info | References
Rendered fromSmCCNet_Vignette_SingleOmics.Rmdusingknitr::rmarkdown

Last update: 2026-05-08
Started: 2023-10-03