SCSilicon

SCSilicon generates synthetic single-cell DNA sequencing data to simulate genomic aberrations for benchmarking and validation of computational methods in cell-specific cancer genomics.

Key Features:

• Automated generation of genomic aberrations: Produces synthetic DNA reads simulating Single Nucleotide Polymorphisms (SNPs), Single Nucleotide Variants (SNVs), Insertions and Deletions (Indels), and Copy Number Variations (CNVs).
• Ground truth CNV segmentation and subclone labels: Provides ground truth CNV segmentation breakpoints and subclone cell labels for validation of CNV callers.
• Benchmarking dataset generation: Generates controlled and scalable volumes of synthetic single-cell data to enable robust benchmarking of computational tools.

Scientific Applications:

• Cancer genomics benchmarking: Enables quantitative evaluation of algorithms that detect cell-specific genomic aberrations in cancer using single-cell DNA sequencing data.
• Development and validation of CNV callers: Provides ground truth datasets and synthetic reads to train, test, and compare single-cell CNV detection methods.
• Algorithm testing for single-cell variant detection: Facilitates assessment of methods for detecting SNPs, SNVs, Indels, and CNVs at single-cell resolution.

Methodology:

Automatically creates synthetic single-cell genomes with predefined aberrations and generates corresponding synthetic DNA reads simulating SNPs, SNVs, Indels, and CNVs, while producing ground truth CNV segmentation breakpoints and subclone cell labels in a controlled, scalable manner.