Package: mgcca 0.99.7

Dolors Pelegri-Siso

mgcca: Generalized Canonical Correlation Analysis with Missing Individuals

Implements Generalized Canonical Correlation Analysis (GCCA) with missing individuals (whole missing rows; van de Velden and Bijmolt, 2006) for multi-block omics integration. The package operates directly on HDF5 matrices via the BigDataStatMeth C++ API to scale computations; R functions are thin wrappers around the C++ core.

Authors:Dolors Pelegri-Siso [aut, cre], Isaac Subirana [aut], Juan R. Gonzalez [aut]

mgcca_0.99.7.tar.gz
mgcca_0.99.7.zip(r-4.6-x86_64)
mgcca_0.99.7.tgz(r-4.6-x86_64)mgcca_0.99.7.tgz(r-4.6-arm64)mgcca_0.99.7.tgz(r-4.5-x86_64)mgcca_0.99.7.tgz(r-4.5-arm64)
mgcca_0.99.7.tar.gz(r-4.7-arm64)mgcca_0.99.7.tar.gz(r-4.7-x86_64)mgcca_0.99.7.tar.gz(r-4.6-arm64)mgcca_0.99.7.tar.gz(r-4.6-x86_64)
manual.pdf |manual.html✨
DESCRIPTION |NEWS
card.svg |card.png
mgcca/json (API)

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

Bug tracker:https://github.com/isglobal-brge/mgcca/issues

Uses libs:
  • openblas– Optimized BLAS
  • curl– Easy-to-use client-side URL transfer library
  • openssl– Secure Sockets Layer toolkit
  • c++– GNU Standard C++ Library v3
  • openmp– GCC OpenMP (GOMP) support library
Datasets:
  • X1 - Simulated cardiovascular multi-table example data
  • X2 - Simulated cardiovascular multi-table example data
  • X3 - Simulated cardiovascular multi-table example data

On CRAN:

Conda:

statisticalmethoddimensionreductionprincipalcomponentgeneexpressiontranscriptomicsepigeneticsmetabolomicsopenblascurlopensslcppopenmp

4.30 score 1 stars 7 scripts 29 exports 54 dependencies

Last updated from:ebf1066760. Checks:1 WARNING, 10 OK, 5 FAIL. Indexed: yes.

TargetResultTimeFilesSyslog
bioc-checksWARNING1209
linux-devel-arm64OK1479
linux-devel-x86_64OK2034
source / vignettesOK2914
linux-release-arm64OK2402
linux-release-x86_64OK3281
macos-release-arm64OK1551
macos-release-x86_64OK2291
macos-oldrel-arm64OK1362
macos-oldrel-x86_64OK2572
windows-devel-arm64FAIL2119
windows-devel-x86_64FAIL3648
windows-release-arm64FAIL2104
windows-release-x86_64OK3529
windows-oldrel-x86_64FAIL3641
wasm-releaseFAIL3665

Exports:corTestPgenCoordsgetSignifgetTablesmatrixChr2Nummgccamgcca_associatemgcca_auditmgcca_audit_loadmgcca_block_profilemgcca_import_hdf5mgcca_lambda_scale_hdf5mgcca_loadmgcca_permtestmgcca_resultsmgcca_select_lambdamgcca_sensitivitymgcca_stabilitymgccaImputeplotAVEplotBiplotplotIndividualsplotIndsplotLoadingsplotScoresplotScreeplotVariablesplotVarstopVars

Dependencies:abindBHBigDataStatMethBiobaseBiocGenericsbiocmakebitopsclicpp11data.tableDelayedArraydir.expiryfarverfilelockgenericsGenomicRangesggplot2ggthemesgluegtableimputeIRangesisobandlabelinglatticelifecyclemagrittrMASSMatrixMatrixGenericsmatrixStatspillarpkgconfigpurrrR6RColorBrewerRcppRcppEigenRCurlRhdf5librlangS4ArraysS4VectorsS7scalesSeqinfoSparseArraySummarizedExperimenttibbleutf8vctrsviridisLitewithrXVector

How much of a shared subspace is carried by group structure?
Introduction | Reading the numbers without the algebra | Two words used throughout | What "alignment" actually means | The four numbers | In what order to look at things | What each combination means | Installation | A self-contained example | Does the omic signal know anything about my variable? | Which data type is the finding coming from? | When your study has batches, centres or stages | Is my result leaning on those groups? | Seeing it instead of inferring it | Which participants is it coming from? | Would I get the same answer with a different sample? | A result you can lean on | A result you cannot | What made the difference | Practical matters | Where the computation happens | Keeping your results | What you can and cannot claim | Session information
Rendered frommgcca_reliability.Rmdusingknitr::rmarkdown

Last update: 2026-10-24
Started: 2026-09-08

Generalized Canonical Correlation with missing individuals - Examples
Introduction | Related packages | Installation | A self-contained example | Running mgcca | Individuals plot | Variables plot | Variance explained | Significant variables | The two plotting families | Saving and reloading results | How it scales to omics data | A live multi-omics example: miniACC | Assembling the tables | Choosing the penalty | Integrating the two layers | Reading the components | Are the components themselves significant? | Scoring individuals the fit never saw | Auditing the availability design before fitting | Auditing the object you are about to fit | Appendix: the full-scale TCGA walkthrough (not evaluated) | Download, impute and assemble the tables | Integrate the two layers | Do the shared components track the disease? | What drives the components | Subsetting by genomic ranges | Session information
Rendered frommgcca_example.Rmdusingknitr::rmarkdown

Last update: 2026-09-29
Started: 2026-09-15

Readme and manuals

Help Manual

Help pageTopics
Simulated cardiovascular multi-table example datacardiovascular X1 X2 X3
Two-sided p-value of a correlation coefficientcorTestP
Subset a MultiAssayExperiment object by specific genomic coordinatesgenCoords
Variables significantly correlated with the shared componentsgetSignif
Splits a MultiAssayExperiment in a list of tablesgetTables
Convert a character table in a list of tables to numericmatrixChr2Num
Generalized Canonical Correlation Analysis with missing individualsmgcca
Associate the shared components with external phenotypesmgcca_associate
Pre-fit availability and design auditmgcca_audit print.mgcca_audit summary.mgcca_audit
Read a stored availability audit back from HDF5mgcca_audit_load
Per-block contribution profile of a sensitivity resultmgcca_block_profile
Import data of any supported type into an HDF5 file for mgccamgcca_import_hdf5
Scale-matched block scale from HDF5, over column chunksmgcca_lambda_scale_hdf5
Reload an mgcca result straight from its HDF5 filemgcca_load
Permutation test for the significance of the shared componentsmgcca_permtest
Collect mgcca results from HDF5 into an in-memory 'mgcca' objectmgcca_results
Choose the ridge penalty by held-out cross-block agreementmgcca_select_lambda
Sensitivity of the shared subspace to a participant-level querymgcca_sensitivity print.mgcca_sensitivity summary.mgcca_sensitivity
Stability of the shared subspace under resamplingmgcca_stability plot.mgcca_stability print.mgcca_stability summary.mgcca_stability
Imputes each assay matrix data from a MultiAssayExperiment objectmgccaImpute
Plot an mgcca objectplot.mgcca
Plot an availability auditplot.mgcca_audit
Plot a block contribution profileplot.mgcca_block_profile
Plot a sensitivity resultplot.mgcca_sensitivity
Barplot of the Average Variance Explained (AVE) per tableplotAVE
Biplot: individuals and variable loadings togetherplotBiplot
Plot individuals on the shared componentsplotIndividuals
Base-R scatter plot of the individuals on the shared componentsplotInds
Barplot of the top variable loadings for one componentplotLoadings
Scatter of one table's canonical scoresplotScores
Scree plot of the shared-component eigenvaluesplotScree
Correlation circle of variables for one tableplotVariables
Base-R correlation plots of the variables, one panel per tableplotVars
Project new individuals onto the shared canonical spacepredict.mgcca
Print method for 'ListMAE' objectsprint.ListMAE
Print an mgcca objectprint.mgcca
Print a block contribution profileprint.mgcca_block_profile
Print a lambda selectionprint.mgcca_lambda
Summarise an mgcca objectsummary.mgcca
Variables most correlated with a shared componenttopVars