Preserving friendships in school contacts: An algorithm to construct synthetic temporal networks for epidemic...

Area of Science:

• Epidemiology
• Computational Biology
• Network Science

Background:

• Accurate infectious disease transmission models require high-resolution temporal contact data.
• Existing contact datasets are often short-term, necessitating synthetic data extension for longer epidemic simulations.
• Simple data repetition creates unrealistic transmission chains by ignoring contact renewal.

Purpose of the Study:

• To develop an algorithm for longitudinally extending short-term contact data in a school setting.
• To incorporate the dual nature of contacts, including repeated contacts due to friendships.
• To simulate SARS-CoV-2 spread using the extended data and compare it with simpler methods.

Main Methods:

• Proposed a novel algorithm to extend contact data, preserving friendship structures.
• Utilized an agent-based model for simulating SARS-CoV-2 transmission within a school setting.
• Compared simulation outcomes from friendship-preserving data extension with those from simpler methods.

Main Results:

• Preserving friendships in contact data extension impacts individual infection pathways but not inter-class transmission routes.
• Simulations indicate that two days of contact data collection can generate realistic long-term synthetic contact sequences.
• The proposed algorithm allows for leveraging existing contact data for more robust disease modeling.

Conclusions:

• The developed algorithm provides a more realistic approach to extending contact data for disease transmission modeling.
• Friendship preservation is crucial for accurately modeling individual-level infection dynamics.
• The findings support efficient contact data collection strategies for future epidemiological studies.