spiky

spiky enhances quantitative normalization of cfMeDIP-seq data using synthetic spike-in controls to improve sensitivity and reproducibility of cell‑free methylome quantification for applications including circulating tumor DNA (ctDNA) analysis.

Key Features:

• Synthetic Spike-In Controls: A set of 54 synthetic spike-in DNA fragments with varied methylation status, fragment length, G+C content, and CpG dinucleotide fraction that do not align to the human genome, used as reference controls for quantitative normalization in cfMeDIP-seq.
• Normalization and Bias Correction: Accounts for technical variation in enrichment efficiency due to biophysical properties of DNA fragments and integrates unique molecular indices (UMIs) to control for differential amplification after enrichment.
• Quantitative Modeling: Uses a generalized linear model to convert read counts into absolute quantities of methylated DNA fragments in picomoles while adjusting for fragment length, G+C content, and CpG fraction.
• Batch Effect Mitigation: Reduces batch-associated variance to ≤ 1% of total variance within datasets to enhance reproducibility across experiments.

Scientific Applications:

• Cancer Research: Enables robust quantitative comparisons of DNA methylation profiles in ctDNA to support identification of tumor tissue of origin.
• Prognostic Analysis: Detects methylation signatures in cell-free DNA to inform disease prognosis and monitor potential relapse.
• Epigenetic Studies: Provides precise normalization for comprehensive epigenetic profiling across conditions using cfMeDIP-seq.

Methodology:

Computational methods include use of 54 non‑human-aligning synthetic spike-ins, integration of unique molecular indices (UMIs) to correct for amplification bias, and application of a generalized linear model that converts read counts to picomoles while adjusting for fragment length, G+C content, and CpG fraction to account for enrichment-efficiency variation.