Logo
ScienceStack
Table of Contents
Fetching ...
Sign In

Unlocking the Power of Multi-institutional Data: Integrating and Harmonizing Genomic Data Across Institutions

Yuan Chen, Ronglai Shen, Xiwen Feng, Katherine Panageas

TL;DR

The Bridge model uses a quantile-matched latent variable approach to derive integrated features to preserve information beyond common genes and maximize the utilization of all available data, while leveraging information sharing to enhance both learning efficiency and the model’s capacity to generalize.

Abstract

Cancer is a complex disease driven by genomic alterations, and tumor sequencing is becoming a mainstay of clinical care for cancer patients. The emergence of multi-institution sequencing data presents a powerful resource for learning real-world evidence to enhance precision oncology. GENIE BPC, led by the American Association for Cancer Research, establishes a unique database linking genomic data with clinical information for patients treated at multiple cancer centers. However, leveraging such multi-institutional sequencing data presents significant challenges. Variations in gene panels result in loss of information when the analysis is conducted on common gene sets. Additionally, differences in sequencing techniques and patient heterogeneity across institutions add complexity. High data dimensionality, sparse gene mutation patterns, and weak signals at the individual gene level further complicate matters. Motivated by these real-world challenges, we introduce the Bridge model. It uses a quantile-matched latent variable approach to derive integrated features to preserve information beyond common genes and maximize the utilization of all available data while leveraging information sharing to enhance both learning efficiency and the model's capacity to generalize. By extracting harmonized and noise-reduced lower-dimensional latent variables, the true mutation pattern unique to each individual is captured. We assess the model's performance and parameter estimation through extensive simulation studies. The extracted latent features from the Bridge model consistently excel in predicting patient survival across six cancer types in GENIE BPC data.

Unlocking the Power of Multi-institutional Data: Integrating and Harmonizing Genomic Data Across Institutions

TL;DR

The Bridge model uses a quantile-matched latent variable approach to derive integrated features to preserve information beyond common genes and maximize the utilization of all available data, while leveraging information sharing to enhance both learning efficiency and the model’s capacity to generalize.

Abstract

Cancer is a complex disease driven by genomic alterations, and tumor sequencing is becoming a mainstay of clinical care for cancer patients. The emergence of multi-institution sequencing data presents a powerful resource for learning real-world evidence to enhance precision oncology. GENIE BPC, led by the American Association for Cancer Research, establishes a unique database linking genomic data with clinical information for patients treated at multiple cancer centers. However, leveraging such multi-institutional sequencing data presents significant challenges. Variations in gene panels result in loss of information when the analysis is conducted on common gene sets. Additionally, differences in sequencing techniques and patient heterogeneity across institutions add complexity. High data dimensionality, sparse gene mutation patterns, and weak signals at the individual gene level further complicate matters. Motivated by these real-world challenges, we introduce the Bridge model. It uses a quantile-matched latent variable approach to derive integrated features to preserve information beyond common genes and maximize the utilization of all available data while leveraging information sharing to enhance both learning efficiency and the model's capacity to generalize. By extracting harmonized and noise-reduced lower-dimensional latent variables, the true mutation pattern unique to each individual is captured. We assess the model's performance and parameter estimation through extensive simulation studies. The extracted latent features from the Bridge model consistently excel in predicting patient survival across six cancer types in GENIE BPC data.
Paper Structure (13 sections, 5 equations, 5 figures)

This paper contains 13 sections, 5 equations, 5 figures.

Table of Contents

  1. Introduction
  2. Method
  3. Model framework
  4. Model estimation
  5. Objective function
  6. Identifiability
  7. Model fitting
  8. Generalization to future population
  9. Simulation studies
  10. Simulation settings
  11. Simulation results
  12. Application to the GENIE BPC Study
  13. Discussion

Figures (5)

  • Figure 1: Left panel: Venn diagrams illustrating the limited common genes (among genes with at least one observed mutation) covered by selected sequencing panels across institutions. Right panel: Mutation frequency by institution (indicated by the color of the bar) for the top 20 common genes of each cancer type in GENIE BPC.
  • Figure 2: Distribution of variant allele frequency (VAF) by institution for each cancer type in GENIE BPC. VAF is defined as the number of variant alleles divided by the total number of sequenced alleles.
  • Figure 3: RMSE for outcome prediction on the test datasets under different settings
  • Figure 4: MSEs of each set of parameters across varying training sample sizes
  • Figure 5: Concordance index for time to overall survival from metastasis on the test sets from 100 replications for each cancer type in GENIE BPC