Embedded Files
Welcome to FORM-LNP
FORM-LNP is a platform for predictive modeling of RNA-loaded lipid nanoparticles (LNPs). We aim to provide a user-friendly, science-backed toolkit that helps researchers and developers explore key formulation parameters. Currently, the model is tailored for siRNA–LNPs, and we are actively working to generalize it for a broader range of nucleic acid cargos and formulation conditions.
(Users only need a Google account to access the Google Colab code; no prior coding experience or software installation is required. )
How it Works
How it Works
This tool mimics the experimental process of lipid nanoparticle formation using a two-stage modeling pipeline. We first run molecular dynamics simulations to capture early self-assembly, then use kinetic Monte Carlo simulations to model long-timescale behaviors, accounting for PEGylated lipids' steric repulsion, diffusion, electrostatics, and van der Waals interactions. A machine learning model trained on these simulations predicts LNP size and RNA payload distributions based on formulation parameters, such as flow rate, PEG size, and salt concentration. The trained model is integrated into the tool, allowing users to directly predict outcomes without running any simulations.
LNP Size, Payload Distribution Prediction
Use input formulation conditions and instantly predict LNP characteristics (final LNP size, ratio of empty LNPs, and siRNA payload distribution variance) in turbulent flow mixing. Designed to support formulations in the 10–150 mM salt, 1000–5000 Da PEG, and 1–3% PEG/lipid ratio range.
You can open the code and run it directly to predict properties. This code automatically loads the trained model and performs all necessary operations. (It may take 1/2 minute to download and read the required files)
We’ve also included all the necessary code in this downloadable package. No installation or environment setup is required; it runs directly in Google Colab.
After downloading, unzip the folder and open the README file for setup steps.
Kinetic Modeling
This module enables users to explore the time evolution and mechanistic basis of LNP growth and empty particle formation using kinetic Monte Carlo (kMC) simulations.
The code enables users to simulate the evolution of size and the empty LNP fraction over time, and adjust key parameters such as initial LNP size, PEG molecular weight, and salt concentration.
You can open and run the code directly. This code automatically reads the trained charge regulation model and performs kMC.
Contact: mpial1@jh.edu
Page updated
Google Sites
Report abuse