This program mainly simulates the growth of individual glands with a stochastic birth-death branching process to learn copy number alteration (CNA) rates and selection strengths.
The major output are summary statistics extracted from copy number profiles (CNPs) of multiple glands sampled from a patient. Only copy number gain and loss are simulated. These summary statistics can be used for inference with ABC (approximate Bayesian computation) approach.
There is also a program to simulate copy numbers obtained from multi-region sequencing data of a patient (not well documented for now).
Download the source code and then run make in folder "src" to generate the simulation program 'simgland'.
Comment out relevant line in 'makefile' to generate the simulation program 'simcrc'.
Please run ./simgland -h for all options.
There are three modes of simulation.
- Mode 0 (default): simulating each gland as a node. The simulation starts from a single gland with "optimal" karyotype and stops when there are N glands. When using option -g, a subset of glands will be sampled from both sides of the lineage tree.
- Mode 1: simulating each cell as a node and a gland is composed of a group of cells. The gland will divide into two glands when reaching a certain number of cells.
- Mode 2: simulating each gland as a node. The simulation starts from a single gland with diploid karyotype and stops when there are N glands. To reach "optimal" karyotype quickly, it is recommended to simulate CNAs at chr-level.
In each mode, there are two models of evolution.
- Model 0 (default): neutral.
- Model 1: selection. Under this model, the selection strengths can be specified by option -f.
Option "verbose" controls how much to output.
- When verbose = 0, a vector of summary statistics is written to the standard output. The summary statistics represents
- percentage of genome altered (PGA) relative to start karyotype,
- average pairwise divergence (proportion of different bins that are altered),
- variance of pairwise divergences,
- number of unique breakpoints across all sampled glands,
- number of unique mutations across all sampled glands, and
- average pairwise mismatch (numbers of different mutations).
- When verbose > 0, more information will be written to the standard output and there is a file containing the copy numbers for each sampled gland. Using '--fdeme FILENAME' will output the lineages of glands.
The example commands show the basic usage of the program. They will simulate 1000 glands (starting from "optimal" karyotype) with new CNAs under mutation rate 0.1 and then sample 30 glands from each side of the lineage tree . Please go to folder 'bin' before running the commands.
- Assuming neutral evolution:
./bin/simgland -o ./ -n 1000 -g "2 30 30" --seed $RANDOM --verbose 0 -r 0.1
- Assuming (negative) selection:
./bin/simgland -o ./ -n 1000 -g "2 30 30" --seed $RANDOM --verbose 0 -r 0.1 --model 1 -f 2
Here, Option '--model 1' indicates selection is imposed. Option '--fitness 2' specifies the strengths of selection.