BITES is a scientific software application developed in the Di-Poï Lab (University of Helsinki) in 2026 for exploring and optimizing vertebrate tooth morphologies in computational simulations.
A pre-built Windows executable is also available from the Releases page (see Installation below).
The program interfaces with the morphodynamic tooth development model originally proposed by Salazar-Ciudad and Jernvall 1 and provides tools for automatically identifying parameter combinations that generate specific tooth shapes. Given a target morphology (e.g., a 3D scan of a real tooth), BITES searches the model’s parameter space for cellular and developmental parameter sets capable of producing comparable structures. Instead of manually testing thousands of parameter combinations, BITES uses Bayesian optimization via the Optuna framework to intelligently search for the best parameter configurations.
The software is described and applied to squamate dentitions in Razmadze et al., 2026 (doi: 10.64898/2026.03.12.711251).
See the User's Guide for instructions on how to install and use BITES effectively.
Developmental models are typically used in a forward exploratory manner. BITES enables inverse inference of developmental parameter landscapes from empirical 3D organ morphology, bridging morphogenesis and evolutionary inference.
- Inverse Bayesian optimization of morphodynamic model parameters
- Automated alignment (STatac + CPD)
- Chamfer Distance–based 3D shape comparison
- Multi-stage simulation evaluation
- Parallel optimization support via PostgreSQL
- GUI-based configuration and visualization
- Optuna optimizer proposes a set of tooth model parameters (e.g. growth rates, signaling molecule diffusion coefficients)
- The tooth model morphodynamic simulation runs with those parameters, producing simulated tooth shape(s)
- The simulated shapes are compared to the target shape
- The resulting loss score (shape similarity) is fed back to the optimizer
- Repeat for hundreds or thousands of trials to find the best parameter configurations
Currently supported platforms: Windows
Express installation guide below. For more detailed instructions, see the User's Guide.
A pre-built Windows executable is available from the Releases page:
- Download and extract the zip file
- Run
BITES.exefrom the extracted folder
PostgreSQL is needed for parallel optimization (running multiple trials at once).
- Download the PostgreSQL installer for Windows from https://www.postgresql.org/download/windows/
- Run the installer. Use the default settings (port 5432, superuser
postgres), but deselect Stack Builder in the component selection step - Choose a password during setup — you will need it when configuring BITES
- In BITES, go to the PostgreSQL configuration and enter your credentials (see User's Guide)
If you use this software in academic work, please cite:
Razmadze, D., Ikola, J., Aalto, I.M. & Di-Poï, N. 2026. An alternative patterning mechanism for vertebrate tooth complexity. bioRxiv doi: 10.64898/2026.03.12.711251
This project is licensed under the GNU General Public License v3.0 — see the LICENSE file for details.
Third-party library licenses are listed in THIRD_PARTY_LICENSES.
The morphodynamic tooth model used by BITES is by Salazar-Ciudad and Jernvall 1, with the model executable derived from ToothMaker. Parameter optimization is performed using the Optuna framework 2. Shape comparison uses Chamfer Distance as implemented in point-cloud-utils 3 and the Coherent Point Drift algorithm 4 as implemented in 5.
Footnotes
-
Salazar-Ciudad, I., Jernvall, J., 2010. A computational model of teeth and the developmental origins of morphological variation. Nature, 464, 583–586. doi:10.1038/nature08838 ↩ ↩2
-
Akiba, T., Sano, S., Yanase, T., Ohta, T., Koyama, M., 2019. Optuna: A next-generation hyperparameter optimization framework. Proceedings of the 25th ACM SIGKDD International Conference on Knowledge Discovery & Data Mining, 2623–2631. doi:10.1145/3292500.3330701 ↩
-
Williams, F., 2022. point-cloud-utils (v0.34). github.com/fwilliams/point-cloud-utils ↩
-
Myronenko, A., Song, X., 2010. Point set registration: Coherent point drift. IEEE Transactions on Pattern Analysis and Machine Intelligence, 32(12), 2262–2275. doi:10.1109/TPAMI.2010.46 ↩
-
Gatti, A. A., Khallaghi, S., 2022. PyCPD: Pure NumPy implementation of the coherent PointDrift algorithm. Journal of Open Source Software, 7(80), 4681. doi:10.21105/joss.04681 ↩