Package: CrcBiomeScreen 1.0.0

Chengxin Li

CrcBiomeScreen: An R package for colorectal cancer screening and microbiome analysis

A developed and benchmarked reproducible machine learning framework for microbiome-based colorectal cancer (CRC) screening. By systematically evaluating normalization strategies, taxonomic resolutions, and class imbalance handling. This R package allows users to apply the full pipeline or selectively run specific components depending on their analytical needs. It establishes a scalable foundation for developing interpretable microbiome-based screening tools to support early CRC detection. This approach could be easily implemented in a national screening programme, to improve early detection rates for this disease.

Authors:Chengxin Li [cre, aut], Rishabh Bezbaruah [aut], Henry Wood [aut], Arief Gusnanto [aut]

CrcBiomeScreen_1.0.0.tar.gz
CrcBiomeScreen_1.0.0.zip(r-4.7)CrcBiomeScreen_1.0.0.zip(r-4.6)CrcBiomeScreen_1.0.0.zip(r-4.5)
CrcBiomeScreen_1.0.0.tgz(r-4.6-any)CrcBiomeScreen_1.0.0.tgz(r-4.5-any)
CrcBiomeScreen_1.0.0.tar.gz(r-4.7-any)CrcBiomeScreen_1.0.0.tar.gz(r-4.6-any)
CrcBiomeScreen_1.0.0.tgz(r-4.6-emscripten)
manual.pdf |manual.html✨
card.svg |card.png
CrcBiomeScreen/json (API)
NEWS

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

Bug tracker:https://github.com/omicsforestry/crcbiomescreen/issues

Datasets:

On BioConductor:CrcBiomeScreen-1.1.0(bioc 3.24)CrcBiomeScreen-1.0.0(bioc 3.23)

softwaremicrobiomemetagenomicsclassificationnormalizationvisualization

5.11 score 97 downloads 34 exports 132 dependencies

Last updated from:fa65626d53 (on RELEASE_3_23). Checks:1 NOTE, 9 OK. Indexed: no.

TargetResultTimeFilesSyslog
bioc-checksNOTE297
linux-devel-x86_64OK510
source / vignettesOK419
linux-release-x86_64OK520
macos-release-arm64OK298
macos-oldrel-arm64OK275
windows-develOK1087
windows-releaseOK968
windows-oldrelOK858
wasm-releaseOK269

Exports:checkClassBalanceCreateCrcBiomeScreenObjectCreateCrcBiomeScreenObjectFromTSEEvaluateCrcBiomeScreenEvaluateModelEvaluateRFEvaluateXGBoostFilterDataSetgetAbsoluteAbundancegetModelDatagetModelResultgetNormalizedDatagetOutlierSamplesgetPredictResultgetRelativeAbundancegetSampleDatagetTaxaDatagetTaxaLevelDataKeepTaxonomicLevelLoadTaxaTableModelingRFModelingRF_noweightsModelingXGBoostModelingXGBoost_noweightsNormalizeDataPredictCrcBiomeScreenqcByCmdscaleRunScreeningsetNormalizedData<-setTaxaData<-SplitDataSetSplitTaxasTrainModelsValidateModelOnData

Dependencies:abindapeBHBiobaseBiocGenericsBiocParallelBiostringscaretclasscliclockclueclustercodetoolscpp11crayondata.tableDelayedArraydiagramdigestdirmultdoFuturedoParalleldplyre1071farverfBasicsforeachformatRfsfutile.loggerfutile.optionsfuturefuture.applygenericsGenomicRangesggplot2ggrepelglobalsgluegowergssgtableGUniFrachardhathmsinlineipredIRangesisobanditeratorsjsonliteKernSmoothlabelinglambda.rlatticelavalazyevallifecyclelistenvlubridatemagrittrMASSMatrixMatrixGenericsmatrixStatsmgcvmodeestModelMetricsnlmennetnumDerivparallellypermutepillarpkgconfigplyrprettyunitspROCprodlimprogressprogressrproxypurrrR6rangerrappdirsRColorBrewerRcppRcppEigenrecipesreshape2rlangrmutilrpartS4ArraysS4VectorsS7scalesSeqinfoshapeSingleCellExperimentsnowSparseArraysparsevctrsspatialSQUAREMstablestablediststatipstatmodstringistringrSummarizedExperimentsurvivaltibbletidyrtidyselecttidytreetimechangetimeDatetimeSeriestreeioTreeSummarizedExperimenttzdbutf8vctrsveganviridisLitewithrXVectoryulab.utils

CrcBiomeScreen: Colorectal Cancer Microbiome Screening and Analysis

Chengxin Li

Rendered fromCrcBiomeScreen.Rmdusingknitr::rmarkdownon Jun 04 2026.

Last update: 2026-04-14
Started: 2025-09-03

Readme and manuals

Help Manual

Help pageTopics
Check the sample distribution of the dataset and give the suggestion if need the class weight or notcheckClassBalance
CrcBiomeScreen ClassCrcBiomeScreen-class
Create a CrcBiomeScreen S4 object for microbiome-based CRC analysisCreateCrcBiomeScreenObject
Create a CrcBiomeScreen object from TreeSummarizedExperimentCreateCrcBiomeScreenObjectFromTSE
Evaluate the performance of model predictionsEvaluateCrcBiomeScreen
Evaluate the model to select the optimal modelEvaluateModel
Evaluate the Random Forest modelEvaluateRF
Evaluate the XGBoost modelEvaluateXGBoost
Filter the CrcBiomeScreenObject dataset based on a specific labelFilterDataSet
Accessor for AbsoluteAbundance slot of CrcBiomeScreen objectgetAbsoluteAbundance getAbsoluteAbundance,CrcBiomeScreen-method
Accessor for ModelData slot of CrcBiomeScreen objectgetModelData
Accessor for ModelResult slot of CrcBiomeScreen objectgetModelResult getModelResult,CrcBiomeScreen-method
Accessor for NormalizedData slot of CrcBiomeScreen objectgetNormalizedData getNormalizedData,CrcBiomeScreen-method
Accessor for OutlierSamplesgetOutlierSamples getOutlierSamples,CrcBiomeScreen-method
Accessor for PredictResult slot of CrcBiomeScreen objectgetPredictResult getPredictResult,CrcBiomeScreen-method
Accessor for RelativeAbundance slot of CrcBiomeScreen objectgetRelativeAbundance getRelativeAbundance,CrcBiomeScreen-method
Accessor for SampleData slot of CrcBiomeScreen objectgetModelData,CrcBiomeScreen-method getSampleData getSampleData,CrcBiomeScreen-method
Accessor for TaxaData slot of CrcBiomeScreen objectgetTaxaData getTaxaData,CrcBiomeScreen-method
Accessor for TaxaLevelData slotgetTaxaLevelData getTaxaLevelData,CrcBiomeScreen-method
Summarize abundance data at a given taxonomic levelKeepTaxonomicLevel
Load a custom taxa table for ASV/OTU dataLoadTaxaTable
The packaging function for Random Forest modelingModelingRF
The function for modeling random forest without using class weightsModelingRF_noweights
The packaging function for XGBoost modelingModelingXGBoost
The packaging function for XGBoost modeling without using class weightsModelingXGBoost_noweights
NHSBCSP screening datasetNHSBCSP_screeningData
Normalise the absolute data to relative data by using Total Sum Scaling and Geometric Mean of Pairwise Ratios (GMPR)NormalizeData
Predict the class and probabilities for new dataPredictCrcBiomeScreen
Quality control using classical MDS and outlier detectionqcByCmdscale
Run the screening process for the microbiome dataRunScreening
setNormalizedData<-: Setter for NormalizedData slot of CrcBiomeScreen objectsetNormalizedData-setter setNormalizedData<- setNormalizedData<-,CrcBiomeScreen-method
setTaxaData<-: Setter for TaxaData slot of CrcBiomeScreen objectsetTaxaData-setter setTaxaData<- setTaxaData<-,CrcBiomeScreen-method
Split the dataset into training and test setsSplitDataSet
Split and clean taxonomy stringsSplitTaxas
Thomas 2018 relative abundance datasetThomas_2018_RelativeAbundance
Train the different modelsTrainModels
Predict the validation data by using the trained model in CrcBiomeScreenObjectValidateModelOnData
Zeller 2014 relative abundance datasetZellerG_2014_RelativeAbundance