Related Experiment Video
Updated: May 10, 2026
12:37
Efficient Nucleic Acid Extraction and 16S rRNA Gene Sequencing for Bacterial Community Characterization
Published on: April 14, 2016
38.2K
FastQTLmapping: an ultra-fast and memory efficient package for mQTL-like analysis.
Xingjian Gao1, Jiarui Li2, Xinxuan Liu3
1National Clinical Research Center of Kidney Diseases, Jinling Hospital, Nanjing, Jiangsu, China.
BMC Bioinformatics
|April 27, 2025
Summary
FastQTLmapping offers a faster, memory-efficient solution for methylation quantitative trait loci (mQTL) analysis. This tool significantly speeds up complex genomic analyses with large datasets and covariates.
Area of Science:
- Computational Genomics
- Bioinformatics
- Statistical Genetics
Background:
- Methylation quantitative trait loci (mQTL) analysis involves high-dimensional genetic and epigenetic data, posing computational challenges.
- Existing methods struggle with the scale and complexity of exhaustive multiple regression analysis, especially with numerous covariates.
Purpose of the Study:
- To develop an ultra-fast and memory-efficient solver for comprehensive multiple regression analysis.
- To address the computational demands of large-scale mQTL-like studies.
Main Methods:
- Developed FastQTLmapping, a precompiled C++ software solution.
- Accelerated performance using Intel MKL and GSL libraries.
- Validated against state-of-the-art methods including MatrixEQTL, FastQTL, and TensorQTL.
Main Results:
- FastQTLmapping achieved an order of magnitude speed improvement over existing methods.
- Demonstrated significant reduction in peak memory usage.
- Completed a large mQTL analysis (3500 individuals, 8M SNPs, 0.8M CpGs, 20 covariates) in 4.5 hours with 13.1 GB peak memory.
Conclusions:
- FastQTLmapping expedites comprehensive mQTL analyses for large genomic datasets with covariates.
- Provides a robust and generic approach for efficient computational genomics.
- Potential to streamline mQTL studies and guide future method development.
