SEK: sparsity exploiting k-mer-based estimation of bacterial community composition

Area of Science:

• Bioinformatics
• Computational Biology
• Genomics

Background:

• Accurate bacterial community composition estimation is crucial for metagenomics.
• High-throughput sequencing data presents challenges due to variable read lengths and noise.
• Existing estimation methods are often computationally intensive.

Purpose of the Study:

• To develop a computationally efficient and robust method for bacterial community composition estimation.
• To address the limitations of current time-consuming estimation techniques.

Main Methods:

• Utilized sparsity-enforcing methods, including compressed sensing, for read assignment.
• Developed a statistical model based on kernel density estimation for accurate read assignment.
• Employed convex optimization and a greedy algorithm for efficient computation.

Main Results:

• The proposed method provides a reasonably fast community composition estimation.
• Demonstrated increased robustness to input data variations compared to related methods.
• A simultaneous assignment of all sample reads to a reference database was achieved.

Conclusions:

• The developed method offers a significant improvement in speed and robustness for metagenomic analysis.
• This approach enhances the accuracy of bacterial community profiling from complex sequencing data.
• The freely available Matlab implementation facilitates broader adoption and application.