Accounting for tumor purity improves cancer subtype classification from DNA methylation data

Area of Science:

• Genomics
• Bioinformatics
• Cancer Research

Background:

• Tumor sample classification is crucial for cancer research and precision medicine.
• Clinical tumor samples are mixtures of cancer and normal cells, leading to mixed signals.
• Tumor purity can bias clustering results if not addressed.

Purpose of the Study:

• To develop a model-based clustering method to infer tumor subtypes.
• To account for tumor purity in DNA methylation microarray data analysis.
• To provide an R function for practical application.

Main Methods:

• Developed a model-based clustering approach.
• Utilized DNA methylation microarray data.
• Created an R function named InfiniumClust within the InfiniumPurify package.

Main Results:

• The developed method successfully infers tumor subtypes considering tumor purity.
• Simulation studies and The Cancer Genome Atlas data analysis showed improved results over existing methods.
• The InfiniumClust R function is available on CRAN.

Conclusions:

• The new method enhances tumor subtype classification accuracy by accounting for tumor purity.
• This approach benefits therapeutic development and precision medicine.
• The InfiniumPurify R package offers a valuable tool for researchers.