Modeling in the Time of COVID-19: Statistical and Rule-based Mesoscale Models.

Area of Science:

• * Computational Biology
• * Structural Biology
• * Biophysics

Background:

• * Accurate 3D modeling of biological entities is crucial for understanding their function.
• * Existing methods can be time-consuming and lack flexibility for incorporating diverse data.
• * Mesoscale modeling requires integrating imaging data with biological knowledge.

Purpose of the Study:

• * To develop a rapid and accurate technique for constructing mesoscale biological models.
• * To enable the integration of 2D microscopy data with rule-based geometric information.
• * To create a flexible platform for revising and refining biological models.

Main Methods:

• * A visual-programming technique combining statistical and rule-based modeling.
• * Determination of statistical properties (shape, spatial distribution) from 2D scans.
• * Interactive 3D visualization for defining spatial relationships and rules.

Main Results:

• * Successful generation of scientifically accurate mesoscale 3D models.
• * Application demonstrated on the SARS-CoV-2 virion ultrastructure.
• * Models incorporate both imaging evidence and defined biological rules.

Conclusions:

• * The new technique offers a rapid, efficient, and reviseable approach to biological modeling.
• * This method facilitates the creation of detailed atomistic models.
• * The developed models can guide future research, particularly for viral structures and disease control.