Data, code and source for "Tile blockers as a simple motif to control self-assembly: kinetics and thermodynamics"

Code and data repository for Constantine Evans, Angel Cervera Roldan, Trent Rogers and Damien Woods,

_Tile blockers as a simple motif to control self-assembly: kinetics and thermodynamics_.

### Organization:

- **theory**: contains some unused musings on analytic solutions for temperature window size.

- **simulations**: contains code for theoretical calculations, simulation runs, and plotting, as well as experimental data and simulation outputs used in the paper.

- **paper**: contains code for the paper, and for figure construction.

### Building the paper:

The paper uses Rust and Python for simulations and plotting, Typst for figure layout, and LaTeX for paper typesetting.

Building the paper can be done in several ways:

- **Build paper using existing simulation data**: Run `make paper` in the root directory

- **Complete build (with simulations)**: Run `make clean-data`, then `make paper` in the root directory

- **Build paper only (self-contained)**: Run `make` in the paper directory

Additional targets in the root directory:

- `make paper-tt` - Build paper using Tectonic

- `make paper-lm` - Build paper using Latexmk

- `make figures` - Generate all figures

- `make clean` - Clean build artifacts (but not simulation outputs)

- `make clean-data` - Clean all data including simulation outputs

- `make dist` - Create a distribution tarball

- `make help` - Show all available targets

### Running simulations:

Simulations can be rerun by running `make simulations` in the root directory.

For more specific simulation runs, you can use:

- `make nuc-temps` - Run nucleation temperature simulations

- `make nuc-mults` - Run nucleation concentration multiplier simulations

- `make growth-temps` - Run growth temperature simulations

- `make growth-mults` - Run growth concentration multiplier simulations

You will likely want to change the settings at the top of `simulations/Makefile` to reduce the number of assemblies run per simulation, the number of temperature/concentration points, and the number of threads used. FFS simulations may also use large amounts of RAM when many are run in parallel; reducing threads to lower than your number of processors may be necessary. Simulations were primarily run on i7ie.48xlarge EC2 instances, which have 192 vCPUs and 1.5 TB of RAM, and took several hours with the (unnecessarily high) settings used.

Most scripts in the simulations directory can also be run individually with different settings.