Related Experiment Video
Updated: Aug 6, 2026
Quantification of Protein Interaction Network Dynamics using Multiplexed Co-Immunoprecipitation
Published on: August 21, 2019
Pre-trained knowledge transfer with application in multisystem inflammatory syndrome subphenotyping in children
Xiaokang Liu1, Naimin Jing2, Yiwen Lu3,4
1Department of Statistics and Data Science, University of Missouri, Columbia, MO, USA.
Insights
This study introduces a novel framework for analyzing Multisystem Inflammatory Syndrome in Children (MIS-C) data across different studies. It effectively addresses data privacy and heterogeneity challenges to identify MIS-C subphenotypes.
Area of Science:
- Pediatric infectious diseases
- Computational biology
- Epidemiology
Background:
- Multisystem Inflammatory Syndrome in Children (MIS-C) is a severe post-COVID-19 condition with diverse clinical presentations.
- Identifying MIS-C subphenotypes is crucial for risk stratification and targeted therapy development.
- Analyzing combined data from multiple studies is ideal but hindered by privacy and heterogeneity issues.
Purpose of the Study:
- To develop a novel framework for analyzing multisite MIS-C data without sharing patient-level information.
- To enable knowledge transfer of pre-trained subphenotypes from large source studies to a target study.
- To account for and explain between-study heterogeneity using study-specific mixing proportions.
Main Methods:
- A one-shot summary statistics-based framework for knowledge transfer.
- Utilizing pre-trained subphenotypes from large-scale source studies.
- Employing study-specific subphenotype mixing proportions to model heterogeneity.
Main Results:
- The proposed framework successfully transfers knowledge of shared subphenotypes across studies.
- Study-specific mixing proportions effectively explain clinical heterogeneity between populations.
- Simulations and MIS-C data analysis confirm the method's benefits.
Conclusions:
- The framework offers a privacy-preserving approach for joint analysis of distributed health data.
- It facilitates the identification of MIS-C subphenotypes even with limited target study data.
- This method can advance understanding and treatment of MIS-C and similar complex pediatric conditions.
Abstract:
Multisystem inflammatory syndrome in children (MIS-C) is a severe post-acute sequela of SARS-CoV-2 infection in children and presents with highly heterogeneous signs and symptoms. Disentangling the clinical manifestations of MIS-C by characterizing its subphenotypes can help identify children at risk for severe outcomes and may help identify more targeted therapies. Characterizing subphenotypes usually requires large sample sizes and can often benefit from combining data from multiple sources. However, joint analysis often faces two major challenges: the prohibition of sharing patient-level data due to privacy concerns and between-study population clinical heterogeneity. To address both challenges, we propose a one-shot summary statistics-based framework to transfer knowledge of the shared subphenotypes pre-trained on large-scale source studies to a target study. Study-specific subphenotype mixing proportions are used to explain between-study heterogeneity. Both real-data guided simulation studies and an application to the MIS-C analysis demonstrate the benefits of our method.