Study - Identifying cancer drivers using methylation patterns
Dysregulation of DNA methylation can adversely impact tumor suppressor genes and developmental regulators, driving progression of cancer.
- Institution
- Stanford University
- Year
- Service
- Bioinformatics
Overview
DNA methylation, the addition of a methyl-group to a CpG site, plays a pivotal role in gene regulation. Aberrant DNA methylation is implicated in various diseases, particularly cancer. With the upsurge of high-throughput data detailing genome-wide methylation profiles across diseases, there is a pressing need for computational tools to discern patterns of aberrant methylation.
MethylMix employs a beta mixture model to delineate subpopulations of samples exhibiting differential DNA methylation compared to normal tissue. MethylMix is implemented in R and leverages The Cancer Genome Atlas (TCGA) to define a baseline relationship between methylation patterns and cancer type to enable further study. As a Canary CREST Intern in 2018, I contributed substantial improvements to the runtime and memory efficacy of MethylMix, achieving over a 10x improvement in speed. Additionally, I implemented initial integration with GenePattern and wrote a comprehensive README for the package's GitHub repository, improving usability and accessibilty of the software. MethylMix is available via Bioconductor.
These advancements in MethylMix are pivotal for the scientific and medical community as they provide a more efficient and user-friendly tool to identify methylation-driven genes, offering insights into the epigenetic mechanisms underlying cancer and other diseases. By enabling a deeper understanding of DNA methylation patterns and their impact on gene expression, MethylMix supports the development of more targeted and personalized therapeutic strategies.
What we did
- Bioinformatics
- Methylation
- Software Development