Cell Type Inference Methods

A public list of computational methods for cell type inference based on genome-scale DNA methylation data

Reference-based methods

Conceptually new reference-based estimation methods

• Constrained projection method by Houseman et al.
  Reference: PMID 22568884
  Implementations: methylSpectrum (R, script), minfi (R, BioC), RnBeads (R, BioC)

• Cell epigenotype-specific (CETS) model
  Reference: PMID 23426267
  Implementation: CETS project on r-forge (R)

• EpiDISH
  Reference: PMID 28517979
  Implementation: EpiDISH (R, Bioc)

Methods for the optimal search of the cell type-specific CpG libraries

• IDOL
  Reference: PMID 26956433
  Implementation: unknown

• MethylCIBERSORT
  Reference: PMID 30104673
  Implementation: R package (Zenodo)

Reference-free methods

Reference-free methods for the correction of differential methylation analysis

• FaST-LMM-EWASHer
  Reference: PMID 24464286
  Implementation: FaST-LMM-EWASHer (R)

• RefFreeEWAS
  Reference: PMID 24451622
  Implementation: RefFreeEWAS (R, CRAN)

• ReFACTor
  Reference: PMID 27018579
  Implementations: ReFACTor (R,MATLAB), GLINT (python)

Reference-free deconvolution methods

• RefFreeCellMix
  Reference: PMID 27358049
  Implementation: RefFreeEWAS (R,CRAN)

• EDec
  Reference: PMID 27851969
  Implementation: EDec (R, GitHub)

• MeDeCom
  Reference: PMID 28340624
  Implementation: MeDeCom (R, GitHub)

Benchmarking studies

• https://www.ncbi.nlm.nih.gov/pubmed/28410574
• https://www.biorxiv.org/content/10.1101/570457v1.abstract
• https://www.ncbi.nlm.nih.gov/pubmed/30571772