ClonalSim: Tumor Clonal Evolution Simulator

ClonalSim is a Bioconductor-compatible R package for simulating tumor clonal evolution with realistic sequencing noise. It generates mutational profiles of heterogeneous tumor samples with hierarchical clonal structure.

Features

Realistic Noise Modeling

ClonalSim implements a two-stage noise model:

1. Biological Noise (Intra-Tumor Heterogeneity)

   • Beta distribution for VAF variation
   • Models spatial/temporal heterogeneity within tumors
   • Configurable concentration parameter

2. Technical Sequencing Noise

   • Depth overdispersion: Negative binomial distribution (GC bias, mappability, PCR artifacts)
   • Stochastic read sampling: Binomial distribution (realistic sampling variation)
   • Base calling errors: Illumina-like error rates (0.1% default)

Simulated Clonal Structure

• Hierarchical evolutionary relationships (e.g., Clone1 → Clone2,3 → Clone4)
• Three mutation types:
  • Founder mutations: present in all subclones
  • Shared mutations: present in clone subgroups
  • Private mutations: specific to individual subclones
• Configurable clone frequencies and mutation counts

Generated Output

The package provides:

1. ClonalSimData S4 object with:

   • Mutation data (True_VAF and observed VAF)
   • Simulation parameters
   • Clonal structure information
   • Metadata (date, version, seed)

2. Visualization functions:

   • VAF density plots
   • VAF scatter plots by mutation type
   • Sequencing depth histograms
   • Clonal matrix heatmaps

3. Export formats:

   • GenomicRanges (GRanges objects)
   • VCF format
   • PyClone input format
   • SciClone input format
   • CSV/data.frame

Installation

From GitHub (Development Version)

# Install devtools if needed
if (!require("devtools")) install.packages("devtools")

# Install ClonalSim
devtools::install_github("gbucci/ClonalSim")

From Bioconductor (When Available)

if (!require("BiocManager", quietly = TRUE))
    install.packages("BiocManager")
BiocManager::install("ClonalSim")

Dependencies

The package requires:

• R >= 4.3.0
• Bioconductor packages: GenomicRanges, IRanges, S4Vectors, VariantAnnotation
• CRAN packages: ggplot2, tidyr

Quick Start

Basic Simulation

library(ClonalSim)

# Run simulation with default parameters
sim <- simulateTumor()

# View summary
sim

# Get detailed statistics
summary(sim)

# Visualize results
plot(sim, type = "vaf_density")

Accessing Results

# Get mutation data
mutations <- getMutations(sim)
head(mutations)

# Get true vs observed VAF
true_vaf <- getTrueVAF(sim)      # Biological truth
observed_vaf <- getObservedVAF(sim)  # With sequencing noise

# Get simulation parameters
params <- getSimParams(sim)
params$subclone_freqs

# Get clonal structure
structure <- getClonalStructure(sim)

Visualization

# VAF density plot (most important)
plot(sim, type = "vaf_density")

# VAF scatter plot by mutation type
plot(sim, type = "vaf_scatter")

# Sequencing depth distribution
plot(sim, type = "depth_histogram")

# Clonal matrix heatmap
plot(sim, type = "clone_matrix")

Common Use Cases

Low Purity Tumor (Normal Cell Contamination)

sim_low_purity <- simulateTumor(
  subclone_freqs = c(0.05, 0.10, 0.15)  # Sum = 0.30 (30% tumor purity)
)

High Coverage Sequencing

sim_deep <- simulateTumor(
  sequencing_noise = list(
    mean_depth = 500,
    depth_dispersion = 100,  # More uniform coverage
    error_rate = 0.0005
  )
)

Low Coverage Exome Sequencing

sim_exome <- simulateTumor(
  sequencing_noise = list(
    mean_depth = 50,
    depth_dispersion = 10,  # More variable coverage
    error_rate = 0.002
  )
)

High Intra-Tumor Heterogeneity

sim_high_het <- simulateTumor(
  subclone_freqs = c(0.05, 0.10, 0.15, 0.20, 0.25, 0.25),
  biological_noise = list(
    enabled = TRUE,
    concentration = 20  # Lower = more heterogeneity
  )
)

Ideal Data (No Noise, for Testing)

sim_ideal <- simulateTumor(
  biological_noise = list(enabled = FALSE),
  sequencing_noise = list(enabled = FALSE)
)

Including Germline Variants

Sequencing data includes both somatic and germline variants. Germline variants (heterozygous diploid) appear at VAF ~0.5 regardless of tumor purity:

# 70% purity tumor with germline contamination
sim_germline <- simulateTumor(
  subclone_freqs = c(0.3, 0.4),  # 70% tumor purity
  germline_variants = list(
    enabled = TRUE,
    n_variants = 100,  # Number of germline SNPs
    vaf_expected = 0.5  # Heterozygous diploid
  )
)

# Germline VAF stays at ~0.5 in both pure and contaminated samples
# This is because germline variants are heterozygous in BOTH tumor and normal cells

Bioconductor Integration

Export to GRanges

library(GenomicRanges)

gr <- toGRanges(sim)
gr

# Subset by chromosome
gr_chr1 <- gr[seqnames(gr) == "chr1"]

Export to VCF

# Get VRanges object
vr <- toVCF(sim, sample_name = "TumorSample")

# Write to VCF file
toVCF(sim, output_file = "simulated_mutations.vcf")

Export for Clonal Deconvolution Tools

# PyClone format
toPyClone(sim, file = "pyclone_input.tsv", sample_id = "sample1")

# SciClone format
toSciClone(sim, file = "sciclone_input.tsv")

# Simple CSV
toDataFrame(sim, file = "mutations.csv")

Custom Clonal Structures

# Define custom evolutionary hierarchy
sim_custom <- simulateTumor(
  subclone_freqs = c(0.1, 0.15, 0.2, 0.25, 0.3),
  n_mut_per_clone = c(30, 40, 50, 40, 30),
  n_mut_shared = list(
    "2 3 4 5" = 20,  # Shared by clones 2,3,4,5
    "3 4 5" = 15,    # Shared by clones 3,4,5
    "4 5" = 10,      # Shared by clones 4,5
    "1 2" = 8        # Shared by clones 1,2
  ),
  n_mut_founder = 15
)

Benchmarking Workflow Example

# 1. Generate ground truth
set.seed(42)
sim <- simulateTumor(
  subclone_freqs = c(0.2, 0.3, 0.5),
  n_mut_per_clone = c(50, 75, 50)
)

# 2. Get true mutations
true_mutations <- getMutations(sim)

# 3. Export for your variant caller
toVCF(sim, output_file = "ground_truth.vcf")

# 4. Run your variant caller on the VCF

# 5. Compare results and calculate metrics
# - Sensitivity: TP / (TP + FN)
# - Precision: TP / (TP + FP)
# - VAF correlation: cor(true_vaf, called_vaf)

Documentation

Vignettes

# View introductory vignette
browseVignettes("ClonalSim")

Function Help

# Main function
?simulateTumor

# Noise model functions
?applyBiologicalNoise
?simulateDepth
?simulateSequencingReads

# Export functions
?toGRanges
?toVCF
?toPyClone

# Accessor functions
?getMutations
?getSimParams

Use Cases

ClonalSim is designed for:

• Benchmarking: Test variant callers and mutation detection pipelines
• Algorithm Development: Develop and test clonal deconvolution algorithms
• Education: Teach tumor heterogeneity and clonal evolution concepts
• Method Validation: Positive controls for analysis pipelines
• Research: Study effects of sequencing parameters on variant detection

Advanced: Standalone Script

For users who prefer the original standalone script (generates all plots automatically):

# The original script is preserved in inst/scripts/
source(system.file("scripts", "simulate_tumor_clones.R", package = "ClonalSim"))

Or directly:

# From the package directory
Rscript inst/scripts/simulate_tumor_clones.R

Output Data Structure

The ClonalSimData object contains:

# Mutations data.frame with columns:
# - Mutation: Unique identifier
# - Chromosome, Position, Ref, Alt: Genomic coordinates
# - True_VAF: Biological truth (with heterogeneity)
# - VAF: Observed VAF (with sequencing noise)
# - Depth: Sequencing coverage
# - Alt_reads: Alternative allele read count
# - Clone: Which clone(s) carry the mutation
# - Type: founder, shared, or private
# - Clone_IDs: Comma-separated clone identifiers

Biological Model

Assumptions

• Diploidy: Assumes diploid genome (max theoretical VAF = 0.5 for heterozygous mutations)
• Clonal Hierarchy: Tree-like phylogenetic structure (no reticulate evolution)
• Heterozygous Mutations: All mutations assumed heterozygous
• No CNAs/LOH: Does not currently simulate Copy Number Alterations or Loss of Heterozygosity

Noise Model Details

Biological Noise (Beta Distribution):

• For clone with frequency f, VAF ~ Beta(α, β)
• α = f × concentration, β = (1-f) × concentration
• Higher concentration → less variability

Technical Noise:

• Depth ~ NegativeBinomial(mean, dispersion)
• Alt_reads ~ Binomial(depth, VAF + error_rate)
• More realistic than Gaussian/Poisson models

Citation

If you use ClonalSim in your research, please cite:

Bucci, G. (2025). ClonalSim: Simulation of Tumor Clonal Evolution with
Realistic Sequencing Noise. https://github.com/gbucci/ClonalSim
DOI: 10.5281/zenodo.17465573

Contributing

Contributions are welcome! Please:

1. Fork the repository
2. Create a feature branch (git checkout -b feature/AmazingFeature)
3. Commit your changes (git commit -m 'Add AmazingFeature')
4. Push to the branch (git push origin feature/AmazingFeature)
5. Open a Pull Request

License

This project is licensed under the MIT License - see the LICENSE file for details.

Authors

Created by Gabriele Bucci for bioinformatics applications in computational oncology.

Support

• Issues: https://github.com/gbucci/ClonalSim/issues
• Documentation: Run browseVignettes("ClonalSim") in R

References

1. McGranahan N, Swanton C. Clonal Heterogeneity and Tumor Evolution: Past, Present, and the Future. Cell. 2017
2. Dentro SC, Wedge DC, Van Loo P. Principles of Reconstructing the Subclonal Architecture of Cancers. Cold Spring Harb Perspect Med. 2017
3. Roth A, et al. PyClone: statistical inference of clonal population structure in cancer. Nat Methods. 2014
4. Miller CA, et al. SciClone: inferring clonal architecture and tracking the spatial and temporal patterns of tumor evolution. PLoS Comput Biol. 2014

Related Tools

• PyClone: Statistical inference of clonal populations
• SciClone: Clonal architecture inference
• ABSOLUTE: Tumor purity and ploidy estimation
• THetA: Copy number and purity inference
• PhylogicNDT: Cancer phylogeny reconstruction

---

Note: ClonalSim generates synthetic data for educational, research, and benchmarking purposes.