Package: netDx 1.17.0

Shraddha Pai

netDx: Network-based patient classifier

netDx is a general-purpose algorithm to build a patient classifier from heterogenous patient data. The method converts data into patient similarity networks at the level of features. Feature selection identifies features of predictive value to each class. Methods are provided for versatile predictor design and performance evaluation using standard measures. netDx natively groups molecular data into pathway-level features and connects with Cytoscape for network visualization of pathway themes. For method details see: Pai et al. (2019). netDx: interpretable patient classification using integrated patient similarity networks. Molecular Systems Biology. 15, e8497

Authors:Shraddha Pai [aut, cre], Philipp Weber [aut], Ahmad Shah [aut], Luca Giudice [aut], Shirley Hui [aut], Anne Nøhr [ctb], Indy Ng [ctb], Ruth Isserlin [aut], Hussam Kaka [aut], Gary Bader [aut]

netDx_1.17.0.tar.gz
netDx_1.17.0.zip(r-4.5)netDx_1.17.0.zip(r-4.4)netDx_1.17.0.zip(r-4.3)
netDx_1.17.0.tgz(r-4.4-any)netDx_1.17.0.tgz(r-4.3-any)
netDx_1.17.0.tar.gz(r-4.5-noble)netDx_1.17.0.tar.gz(r-4.4-noble)
netDx_1.17.0.tgz(r-4.4-emscripten)netDx_1.17.0.tgz(r-4.3-emscripten)
netDx.pdf |netDx.html✨
netDx/json (API)
NEWS

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

Peer review:

Datasets:
  • MB.pheno - Sample metadata table for medulloblastoma dataset.
  • cnv_GR - CNV locations for breast cancer
  • cnv_TTstatus - List of train/test statuses for CNV example
  • cnv_netPass - Vector of pathways that pass class enrichment
  • cnv_netScores - List of pathway-level feature selection scores
  • cnv_patientNetCount - Binary matrix of patient occurrence in networks
  • cnv_pheno - Data.frame of patient labels and status for CNV example
  • confmat - Confusion matrix example
  • featScores - Demo feature-level scores from running feature selection on two-class problem
  • genes - Table of gene definitions
  • modelres - Sample output of getResults
  • npheno - Toy sample metadata table
  • pathwayList - Sample list of pathways
  • pathway_GR - List of genomic ranges mapped to pathways
  • pheno - Sample metadata table
  • pheno_full - Subsample of TCGA breast cancer data used for netDx function examples
  • predRes - Example output of getPatientRankings, used to call labels for test patients.
  • silh - Toy network.
  • toymodel - Example model returned by a buildPredictor() call.
  • xpr - Example expression matrix

On BioConductor:netDx-1.17.0(bioc 3.20)netDx-1.16.0(bioc 3.19)

This package does not link to any Github/Gitlab/R-forge repository. No issue tracker or development information is available.

bioconductor-package

74 exports 1.80 score 102 dependencies 4 mentions

Last updated 2 months agofrom:827f9cec36

Exports:avgNormDiffbuildPredictorbuildPredictor_sparseGeneticcallFeatSelcallOverallSelectedFeaturescleanPathwayNamecompareShortestPathcompileFeaturescompileFeatureScoresconfusionMatrixconvertProfileToNetworksconvertToMAEcountIntTypecountIntType_batchcountPatientsInNetcreateNetFuncFromSimListcreatePSN_MultiDatadataList2ListenrichLabelNetsfetchPathwayDefinitionsgetEMapInputgetEMapInput_manygetEnrgetFeatureScoresgetFileSepgetGMjar_pathgetNetConsensusgetORgetPatientPredictionsgetPatientRankingsgetPerformancegetPSNgetRegionOLgetResultsgetSimilaritymakeInputForEnrichmentMapmakePSN_NamedMatrixmakePSN_RangeSetsmakeQueriesmakeSymmetricmapNamedRangesToSetsnormDiffperfCalcplotEmapplotIntegratedPatientNetworkplotPerfplotPerf_multipredictpredictPatientLabelspruneNetpruneNet_pctXpruneNetsrandAlphanumStringreadPathwaysRR_featureTallyrunFeatureSelectionrunQuerysetupFeatureDBsim.eucscalesim.pearscalesimpleCapsmoothMutations_LabelPropsparsify2sparsify3splitTestTrainsplitTestTrain_resamplingsubsampleValidationDatathresholdSmoothedMutationstSNEPlotterupdateNetswriteNetsSIFwriteQueryBatchFilewriteQueryFilewriteWeightedNets

Dependencies:abindaskpassBHbigmemorybigmemory.sriBiobaseBiocBaseUtilsBiocFileCacheBiocGenericsbitbit64bitopsblobcachemcaToolsclicodetoolscolorspacecombinatcpp11crayoncurlDBIdbplyrDelayedArraydoParalleldplyrfansifarverfastmapfilelockforeachgenericsGenomeInfoDbGenomeInfoDbDataGenomicRangesggplot2glmnetgluegplotsgtablegtoolshttrigraphIRangesisobanditeratorsjsonliteKernSmoothlabelinglatticelifecyclemagrittrMASSMatrixMatrixGenericsmatrixStatsmemoisemgcvmimeMultiAssayExperimentmunsellnlmeopensslpillarpkgconfigplogrplotrixplyrpracmapurrrR6rappdirsRColorBrewerRcppRcppEigenreshape2rlangROCRRSQLiteRtsneS4ArraysS4VectorsscalesshapeSparseArraystringistringrSummarizedExperimentsurvivalsystibbletidyrtidyselectUCSC.utilsutf8uuidvctrsviridisLitewithrXVectorzlibbioc

Readme and manuals

Help Manual

Help pageTopics
wrapper function for getting BiocFileCache associated with netDx package.get_cache
built-in similarity functionsallowedSims
takes average of normdiff of each row in xavgNormDiff
Run nested cross-validation on databuildPredictor
Performs feature selection using multiple resamplings of the databuildPredictor_sparseGenetic
Return feature selected nets based on given criteriacallFeatSel
Wrapper to call selected featurescallOverallSelectedFeatures
internal test function to check validity of makeNetFunc and simscheckMakeNetFuncSims
checks if provided similarity functions are valid. Returns error if notcheckSimValid
Clean pathway name so it can be a filename.cleanPathwayName
CNV locations for breast cancer (subset)cnv_GR
Vector of pathways that pass class enrichmentcnv_netPass
List of pathway-level feature selection scorescnv_netScores
Binary matrix of patient occurrence in networkscnv_patientNetCount
data.frame of patient labels and status for CNV examplecnv_pheno
list of train/test statuses for CNV examplecnv_TTstatus
compare intra-cluster shortest distance to overall shortest distance of the networkcompareShortestPath
Create GeneMANIA databasecompileFeatures
Tally the score of networks through cross-validationcompileFeatureScores
Confusion matrix exampleconfmat
Make confusion matrixconfusionMatrix
Convert profiles to interaction networks before integrationconvertProfileToNetworks
Wrapper that converts an input list into a MultiAssayExperiment objectconvertToMAE
Counts the number of (+,+) and (+,-) interactions in a single networkcountIntType
Counts number of (+,+) and (+,-) interactions in a set of networkscountIntType_batch
Count number of patients in a networkcountPatientsInNet
Create PSN from provided similaritiescreateNetFuncFromSimList
Wrapper to create custom input features (patient similarity networks)createPSN_MultiData
Convert MultiAssayExperiment object to list and data.framedataList2List
Score networks based on their edge bias towards (+,+) interactionsenrichLabelNets
Demo feature-level scores from running feature selection on two-class problemfeatScores
fetch pathway definitions from downloads.baderlab.orgfetchPathwayDefinitions
Table of gene definitions (small subsample of human genes)genes
Counts the relative correlation of (+,+) and (+,-)(-,-) interactionsgetCorrType
write enrichment map for consensus netsgetEMapInput
Wrapper to generate multiple EnrichmentMaps (perhaps one per class)getEMapInput_many
Get ENR for all networks in a specified directorygetEnr
Compile network scores into a matrixgetFeatureScores
platform-specific file separatorgetFileSep
download and update GeneMANIA jar filegetGMjar_path
compile net score across a set of predictor resultsgetNetConsensus
Get relative proportion of patient classes that contribute to a set of networksgetOR
Calculates patient-level classification accuracy across train/test splitsgetPatientPredictions
Process GM PRANK files to get the ROC curve for the querygetPatientRankings
performance metrics for modelgetPerformance
get the integrated patient similarity network made of selected featuresgetPSN
Returns overlapping named ranges for input rangesgetRegionOL
Compiles performance and selected features for a trained model.getResults
Measures of patient similaritygetSimilarity
Wrapper to create input files for Enrichment MapmakeInputForEnrichmentMap
Create patient networks from full matrix of named measurementsmakePSN_NamedMatrix
Create patient similarity interaction networks based on range setsmakePSN_RangeSets
Randomly select patients for queries for feature selectionmakeQueries
Convert a network in source-target-weight format to symmetric matrixmakeSymmetric
Map named ranges to corresponding set of named rangesmapNamedRangesToSets
Converts matrix index (1 to m*n) to row (m) and column (n) numbermatrix_getIJ
Sample metadata table for medulloblastoma dataset.MB.pheno
Sample output of getResults()modelres
moves interaction networks when compiling database for sparse genetic workflowmoveInteractionNets
Similarity metric of normalized differencenormDiff
Toy sample metadata tablenpheno
List of genomic ranges mapped to pathwayspathway_GR
Sample list of pathwayspathwayList
Computes variety of predictor evaluation measures based on the confusion matrixperfCalc
Sample metadata tablepheno
Subsample of TCGA breast cancer data used for netDx function examplespheno_full
Create EnrichmentMap in Cytoscape to visualize predictive pathwaysplotEmap
Visualize integrated patient similarity network based on selected featuresplotIntegratedPatientNetwork
Plots various measures of predictor performance for binary classifiersplotPerf
Plots a set of ROC/PR curves with average.plotPerf_multi
predict patient labelspredict
assign patient class when ranked by multiple GM predictorspredictPatientLabels
Example output of getPatientRankings, used to call labels for test patients.predRes
Prune network by retaining strongest edgespruneNet
Prune network by retaining strongest edgespruneNet_pctX
Prune interaction networks to keep only the networks and patients requestedpruneNets
make PSN for built-in similarity functionspsn__builtIn
wrapper for PSNs using Pearson correlationpsn__corr
make PSN for custom similarity functionspsn__custom
Generate random alphanumerical string of length 10randAlphanumString
Parse GMT file and return pathways as listreadPathways
Replace pattern in all files in dirreplacePattern
Computes positive and negative calls upon changing stringency of feature selected networks (binary networks only)RR_featureTally
Run GeneMANIA cross-validation with a provided subset of networksrunFeatureSelection
Run a queryrunQuery
setup database of features for feature selectionsetupFeatureDB
Toy network.silh
Similarity method. Euclidean distance followed by exponential scalingsim.eucscale
various similarity functions Similarity function: Pearson correlation followed by exponential scalingsim.pearscale
simple capitalizationsimpleCap
This function applies the random walk with restart propagation algorithm to a matrix of patients profilessmoothMutations_LabelProp
cleaner sparsification routinesparsify2
cleaner sparsification routine - faster, matrix-based versionsparsify3
Split samples into train/testsplitTestTrain
Assign train/test labels over several resamplings of the data.splitTestTrain_resampling
Subsample a hold-out set from a larger patient datasetsubsampleValidationData
Apply discretization to the matrix resulted from the propagation on the sparse patient matrixthresholdSmoothedMutations
Example model returned by a buildPredictor() call.toymodel
Plot tSNEtSNEPlotter
Synchronize patient set in sample table and network table.updateNets
write patient networks in Cytoscape's .sif formatwriteNetsSIF
Write batch.txt file required to create GeneMANIA databasewriteQueryBatchFile
Wrapper to write GeneMANIA query filewriteQueryFile
Write an integrated similarity network consisting of selected networks.writeWeightedNets
Example expression matrixxpr