Feb 10, 2026
Optimizing Animal Randomization for In Vivo Experiments
Forked from a private protocol
- Sylvain Carlioz1
- 1randmice
External link: https://randmice.com
Protocol Citation: Sylvain Carlioz 2026. Optimizing Animal Randomization for In Vivo Experiments. protocols.io https://dx.doi.org/10.17504/protocols.io.261ge19bov47/v1
Manuscript citation:
Bakhos Jneid et al., Selective STING stimulation in dendritic cells primes antitumor T cell responses.Sci. Immunol.8,eabn6612(2023).DOI:10.1126/sciimmunol.abn6612
License: This is an open access protocol distributed under the terms of the Creative Commons Attribution License, which permits unrestricted use, distribution, and reproduction in any medium, provided the original author and source are credited
Protocol status: Working
We use this protocol and it's working
Created: February 10, 2026
Last Modified: February 10, 2026
Protocol Integer ID: 242954
Keywords: randomization, mice, animal, animal randomization with randmice, animal randomization, randomization protocol, randmice, animals between the group, number of animal, animal, experiment, randomization, optimizing animal randomization for in vivo experiment, optimizing animal randomization, animals per group, combinations of animal assignment, experimental group, animal assignment, random assignment, laboratory animal, vivo experiment, mice, online tool randmice, testing combination, scale experiment, group, lowest covariate imbalance, distribution with the lowest covariate imbalance
Abstract
This protocol describes a standardized procedure to distribute laboratory animals (mice, rats) into experimental groups using the online tool Randmice (https://randmice.com). The algorithm minimizes inter-group heterogeneity by iteratively testing combinations of animal assignments and selecting the distribution with the lowest covariate imbalance. This approach reduces operator bias, improves reproducibility, and supports compliance with the 3Rs principle (Replacement, Reduction, Refinement) as required by European Directive 2010/63/EU. The tool is optimized for small-scale experiments (N < 12 animals per group), where random assignment leads to high inter-group heterogeneity (Bertsimas et al., 2015).
Troubleshooting
Before randomization
Identify all animals with a unique tag
Measure the covariates to be balanced across groups. Example of common covariates are: body weight, tumor volume (left and/or right flank), blood pressure, any other quantitative variable relevant to the experiment
Run randomization
Login if you have an account. Otherwise, enter your email address to access the form.
You have now access to the form to enter your data
Enter the Number of groups
Choose the Number of iterations for the randomization. 10^6 is standard and run very fast (less than 2min) and generates good results. 10^7 is a bit longer and gives excellent results.
Fill your Own reference if needed
Check/uncheck Compare with original balancing if you want to compare your own randomization with the one from randmice.
Enter your data. The 1st fill is a header that describes each column. It can be any text. Tag must be unique identifier. See instructions link for more information.
Click Randomize when everything is ready.
The randomization will proceed. You can close the window if you want, the report will be sent in your mailbox when it is ready.
In your mailbox, you will find the report.
Below is an example:
You will also receive a CSV file to import if needed
Protocol references
Bakhos Jneid et al., Selective STING stimulation in dendritic cells primes antitumor T cell responses.Sci. Immunol.8,eabn6612(2023).DOI:10.1126/sciimmunol.abn6612