MODMatcher: multi-omics data matcher for integrative genomic analysis.

Area of Science:

• Genomics
• Bioinformatics
• Data Science

Background:

• Large-scale genetic and genomic studies are prone to sample annotation and labeling errors during data generation and management.
• Accurate sample labeling is critical for the integrity of integrative genomic analyses.
• Cis-regulations provide a basis for connecting different types of molecular data.

Purpose of the Study:

• To develop a computational approach for identifying and correcting sample labeling errors in multi-omics data.
• To enhance the reliability of integrative genomic studies by ensuring accurate sample annotation.

Main Methods:

• Development of the Multi-Omics Data Matcher (MODMatcher) computational tool.
• Application of MODMatcher to a large lung genomic study.
• Validation through simulation studies comparing different omics data combinations.
• Demonstration of broad applicability to large genomic datasets like The Cancer Genome Atlas (TCGA).

Main Results:

• MODMatcher effectively identifies and corrects sample labeling errors in multi-omics datasets.
• Application to a lung genomic study more than doubled statistically significant genetic associations and genomic correlations.
• Using three types of omics data yielded more robust results than using two types in simulations.

Conclusions:

• Sample annotation and labeling error inspection is an essential data quality assurance step in genomic studies.
• MODMatcher is a valuable tool for improving the accuracy and robustness of integrative genomic analyses.
• The approach is broadly applicable to large-scale multi-omics datasets, including TCGA data.