Local Promoter Methylation Disorder (LPMD) Algorithm Results for DNMT3A-mutated AML

68 AML patients from the GSE62298 dataset were used to develop an algorithm quantifying local methylation heterogeneity. The algorithm was validated in an independent cohort of 20 AML samples and a TCGA-LAML cohort of 194 patients. The dataset includes a 5-DMDR panel for predicting azacitidine response, achieving an AUC of 0.777.

Use Cases

• Predicting azacitidine treatment response in high-risk MDS/secondary AML patients based on the 5-DMDR panel.
• Analyzing bidirectional epigenetic disruption patterns (aberrant stabilization vs. maintenance failure) associated with DNMT3A mutations.
• Monitoring treatment dynamics by tracking changes in Local Promoter Methylation Disorder (LPMD) scores over time.
• Validating the LPMD algorithm's performance across independent cohorts like TCGA-LAML.

Strengths

• Algorithm validated across multiple cohorts, including an independent WGBS cohort (20 samples) and the TCGA-LAML cohort (n=194).
• Predictive 5-DMDR panel achieved an AUC of 0.777 with 81% sensitivity and 73% specificity.
• Analysis includes longitudinal samples (n=153) for treatment dynamics.

Limitations

• Row count is unknown, which may limit suitability assessment.
• Column-level documentation is absent; field semantics must be inferred after download.
• Dataset size is 21.1 KB, indicating a very limited scope, likely containing summary results rather than raw data.

Provenance

Source

figshare, authored by Xin Liu.

Collection Method

Algorithm development and analysis using public datasets (GSE62298, TCGA-LAML, GSE152710).

Freshness

Last updated 2026-03-18 05:21:23; freshness should be verified.