Tutorial

The easiest way to use Crispulator is via the command line config file. If this format is too constraining, the Custom Simulations has a detailed walk- through of writing a custom simulation where each step can be modified according to need.

Graphical overview

The simulation is laid out in the following manner:

Getting started

First, navigate to the Crispulator directory.

$ julia -e 'import Crispulator; println(normpath(pathof(Crispulator), "..", ".."))'

There should be a YAML file called example_config.yml. Open this is in a text editor and it should look like this

# This is an example configuration file. Whitespace is important.

# Settings pertaining to the library design
library:
    genome:
        num-genes: 500
        num-guides-per-gene: 5
        frac-increasing-genes: 0.02 # fraction of genes with a positive phenotype
        frac-decreasing-genes: 0.1 # fraction of genes with a negative phenotype

    guides:
        crispr-type: CRISPRn # either CRISPRi or CRISPRn
        frac-high-quality: 0.9 # fraction of high quality guides
        mean-high-quality-kd: 0.85 # mean knockdown by a high quality guide (CRISPRi only)

screen:
    type: facs # either facs or growth
    num-runs: 10 # how many independent runs

    representation: # integer value, how much larger are samples than the library
        - transfection: 100
        - selection: 100
        - sequencing: 100

# screen-type specific parameters

    bin-size: 0.25 # size of tail to sample from, must be between 0 and 0.5 (FACS only)
    std-noise: 1 # (FACS only)
    num-bottlenecks: 10 # (Growth only)

This gives access to most dials in the simulation, if something is missing than see Custom Simulations.

Now, lets remove all genes that have a positive phenotype by changing line 8 to 0.0:

        frac-increasing-genes: 0.0 # fraction of genes with a positive phenotype

Running simulation

Now, we can actually run the code by executing the following command

julia run.jl config example_config.yml test_output

The output should look like

[ Info: Activating simulation environment
  Activating project at `~/work/Crispulator.jl/Crispulator.jl`
[ Info: Instantiating environment
  Activating project at `~/work/Crispulator.jl/Crispulator.jl`
[ Info: Loading simulation framework
[ Info: Directory test_output does not exist, attempting to create
[ Info: Using 1 thread(s)
[ Info: Parsing config
[ Info: Running config
[ Info: Analyzing results
[ Info: Saving results in test_output

Quick results:
##############
Venn score = 0.992, 95% conf int (0.979, 1.006)
AUPRC score = 0.917, 95% conf int (0.885, 0.949)
SNR score = 4.055 +/- 0.382

The test_output/ directory should now be populated with all the files

counts.svg
results_table.csv
volcano.svg

Output

The folder contains one of the raw count scatterplots (left) and a volcano plot of mean log2 fold change versus significance of each gene (right)

It also has a useful table that contains all the summary statistic information.

The table below describes each column

Experiments from the paper

This repository also includes a collection of experiments that were run for the paper. You can view the full list by running

julia run.jl ls

They are located in the exps/ directory and also listed here for convenience:

You can run them as follows:

julia run.jl exp growth_sensitivity_library.jl output.csv

where the simulation result will be saved to output.csv.