CytoMethIC is a comprehensive package that provides model data and functions for easily using machine learning models that use data from the DNA methylome to classify cancer type and phenotype from a sample. The primary motivation for the development of this package is to abstract away the granular and accessibility-limiting code required to utilize machine learning models in R. Our package provides this abstraction for RandomForest, e1071 Support Vector, Extreme Gradient Boosting, and Tensorflow models. This is paired with an ExperimentHub component, which contains our lab’s models developed for epigenetic cancer classification and predicting phenotypes. This includes CNS tumor classification, Pan-cancer classification, race prediction, cell of origin classification, and subtype classification models.

MODELS

Models available are listed below:

CytoMethIC Basic Models
EHID ModelID PredictionLabel
NA Age_HM450_20240504 Age prediction (year)
NA Age_HM450_20240611 Age prediction (year)
NA Age_MM285_20220101 Age prediction (year)
NA Age_MM285_20230101 Age prediction (year)
NA CellMethID_mouseBlood_MM285 Deconvolution model for mouse blood components
NA LeukoFrac_HM27_20240614 Leukocyte fraction prediction (%)
NA LeukoFrac_HM450_20240614 Leukocyte fraction prediction (%)
NA MIR200C_EPIC_20240315 Mesenchymal score based on Mir200C meth [0-1]
NA Race3_InfHum3_20240114 Races (N=3)
EH8421 Race5_rfc Races (N=5)
NA Race5_rfcTCGA_InfHum3 Races (N=5)
NA RepliTali_EPIC_20240315 Replication/mitotic age (scale-less)
NA Sex2_HM450_20240114 Sex (N=2)
NA Sex2_MM285_20240114 Sex (N=2)
NA TissueComp_EPIC_20240717 Tissue composition (%)
NA TissueComp_EPICv2_20240717 Tissue composition (%)
NA TissueComp_HM450_20240827 Tissue composition (%)
NA TissueComp_MSA_20240717 Tissue composition (%)
NA TissueType_EPIC_20240610 Dominating tissue type
NA TissueType_EPIC_20240624 Dominating tissue type
NA TissueType_EPICv2_20240716 Dominating tissue type

One can access the model using the EHID above in ExperimentHub()[["EHID"]].

More models (if EHID is NA) are available in the following Github Repo. You can directly download them and load with readRDS(). Some examples using either approach are below.

EXAMPLE INPUT

library(sesame)
library(CytoMethIC)
betasHM450 = imputeBetas(sesameDataGet("HM450.1.TCGA.PAAD")$betas)

To make models work for incompatible platforms, you could try the mLiftOver. Here is an example:

betasEPIC = openSesame(sesameDataGet("EPICv2.8.SigDF")[[1]], mask=FALSE)
betasHM450 = imputeBetas(mLiftOver(betasEPIC, "HM450"))

SEX

model = readRDS(url("https://github.com/zhou-lab/CytoMethIC_models/raw/refs/heads/main/models/Sex2_HM450_20240114.rds"))
cmi_predict(betasHM450, model)
## $score
## [1] 0.8132805
## 
## $sex
## [1] "MALE"

AGE

model = readRDS(url("https://github.com/zhou-lab/CytoMethIC_models/raw/refs/heads/main/models/Age_HM450_20240504.rds"))
cmi_predict(betasHM450, model)
## $age
## [1] 84.13913
## 
## $x
## [1] 3.054244

RACE

The below snippet shows a demonstration of the cmi_predict function working to predict the ethnicity of the patient.

model = ExperimentHub()[["EH8421"]] # the same as "https://github.com/zhou-lab/CytoMethIC_models/raw/refs/heads/main/models/Race5_rfcTCGA_InfHum3.rds"
cmi_predict(betasHM450, model)
## $response
## [1] "WHITE"
## 
## $prob
## WHITE 
## 0.886

CELL FRACTIONS

## leukocyte fractions
model = readRDS(url("https://github.com/zhou-lab/CytoMethIC_models/raw/refs/heads/main/models/LeukoFrac_HM450_20240614.rds"))
cmi_predict(betasHM450, model)
## $leukoFrac
## [1] 0.1960776

Cell-type deconvolution using Loyfer et al. conferences:

model = readRDS(url("https://github.com/zhou-lab/CytoMethIC_models/raw/refs/heads/main/models/TissueComp_HM450_20240827.rds"))
cell_comps = cmi_predict(betasHM450, model)
cell_comps = enframe(cell_comps$frac, name="cell_type", value="frac")
cell_comps = cell_comps |> filter(frac>0)

ggplot(cell_comps, aes(x="", y=frac, fill=cell_type)) +
    geom_bar(stat="identity", width=1) +
    coord_polar(theta="y") +
    theme_void() + labs(fill = "Cell Type") +
    theme(plot.title = element_text(hjust = 0.5))

deconvolution

sessionInfo()
## R version 4.5.0 beta (2025-04-02 r88102)
## Platform: x86_64-pc-linux-gnu
## Running under: Ubuntu 24.04.2 LTS
## 
## Matrix products: default
## BLAS:   /home/biocbuild/bbs-3.22-bioc/R/lib/libRblas.so 
## LAPACK: /usr/lib/x86_64-linux-gnu/lapack/liblapack.so.3.12.0  LAPACK version 3.12.0
## 
## locale:
##  [1] LC_CTYPE=en_US.UTF-8       LC_NUMERIC=C              
##  [3] LC_TIME=en_GB              LC_COLLATE=C              
##  [5] LC_MONETARY=en_US.UTF-8    LC_MESSAGES=en_US.UTF-8   
##  [7] LC_PAPER=en_US.UTF-8       LC_NAME=C                 
##  [9] LC_ADDRESS=C               LC_TELEPHONE=C            
## [11] LC_MEASUREMENT=en_US.UTF-8 LC_IDENTIFICATION=C       
## 
## time zone: America/New_York
## tzcode source: system (glibc)
## 
## attached base packages:
## [1] stats     graphics  grDevices utils     datasets  methods   base     
## 
## other attached packages:
##  [1] sesame_1.27.0        sesameData_1.27.0    CytoMethIC_1.5.0    
##  [4] ExperimentHub_2.17.0 AnnotationHub_3.17.0 BiocFileCache_2.17.0
##  [7] dbplyr_2.5.0         BiocGenerics_0.55.0  generics_0.1.3      
## [10] knitr_1.50          
## 
## loaded via a namespace (and not attached):
##  [1] tidyselect_1.2.1            dplyr_1.1.4                
##  [3] blob_1.2.4                  filelock_1.0.3             
##  [5] Biostrings_2.77.0           fastmap_1.2.0              
##  [7] digest_0.6.37               mime_0.13                  
##  [9] lifecycle_1.0.4             KEGGREST_1.49.0            
## [11] RSQLite_2.3.9               magrittr_2.0.3             
## [13] compiler_4.5.0              rlang_1.1.6                
## [15] sass_0.4.10                 tools_4.5.0                
## [17] yaml_2.3.10                 S4Arrays_1.9.0             
## [19] bit_4.6.0                   curl_6.2.2                 
## [21] DelayedArray_0.35.0         plyr_1.8.9                 
## [23] RColorBrewer_1.1-3          abind_1.4-8                
## [25] BiocParallel_1.43.0         withr_3.0.2                
## [27] purrr_1.0.4                 grid_4.5.0                 
## [29] stats4_4.5.0                preprocessCore_1.71.0      
## [31] wheatmap_0.2.0              colorspace_2.1-1           
## [33] ggplot2_3.5.2               scales_1.3.0               
## [35] SummarizedExperiment_1.39.0 cli_3.6.4                  
## [37] rmarkdown_2.29              crayon_1.5.3               
## [39] reshape2_1.4.4              httr_1.4.7                 
## [41] tzdb_0.5.0                  DBI_1.2.3                  
## [43] cachem_1.1.0                stringr_1.5.1              
## [45] parallel_4.5.0              AnnotationDbi_1.71.0       
## [47] BiocManager_1.30.25         XVector_0.49.0             
## [49] matrixStats_1.5.0           vctrs_0.6.5                
## [51] Matrix_1.7-3                jsonlite_2.0.0             
## [53] IRanges_2.43.0              hms_1.1.3                  
## [55] S4Vectors_0.47.0            bit64_4.6.0-1              
## [57] jquerylib_0.1.4             glue_1.8.0                 
## [59] codetools_0.2-20            stringi_1.8.7              
## [61] gtable_0.3.6                BiocVersion_3.22.0         
## [63] GenomeInfoDb_1.45.0         GenomicRanges_1.61.0       
## [65] UCSC.utils_1.5.0            munsell_0.5.1              
## [67] tibble_3.2.1                pillar_1.10.2              
## [69] rappdirs_0.3.3              htmltools_0.5.8.1          
## [71] randomForest_4.7-1.2        GenomeInfoDbData_1.2.14    
## [73] R6_2.6.1                    evaluate_1.0.3             
## [75] Biobase_2.69.0              lattice_0.22-7             
## [77] readr_2.1.5                 png_0.1-8                  
## [79] memoise_2.0.1               BiocStyle_2.37.0           
## [81] bslib_0.9.0                 Rcpp_1.0.14                
## [83] SparseArray_1.9.0           xfun_0.52                  
## [85] MatrixGenerics_1.21.0       pkgconfig_2.0.3