Identifying Taxonomic Units in Metagenomic DNA Streams on Mobile Devices

Area of Science:

• Bioinformatics
• Computational Biology
• Genomics

Background:

• Portable DNA sequencers like Oxford Nanopore Technology MinION enable real-time, in-field metagenomic sequencing.
• Current metagenomic analysis tools are compute-intensive and designed for batch processing, limiting their use on mobile devices.

Purpose of the Study:

• To develop a memory-efficient approach for identifying Operational Taxonomic Units (OTUs) in metagenomic DNA streams directly on mobile devices.
• To adapt metagenomic analysis for real-time, on-site applications using portable sequencing technology.

Main Methods:

• Proposed a novel algorithm for identifying OTUs by finding connected components in overlap graphs constructed from streaming DNA reads.
• Developed an efficient streaming algorithm to maintain connected components and utilized transitive closures to remove redundant information.
• Integrated the algorithms into a mobile-tailored DNA analysis pipeline.

Main Results:

• Demonstrated high precision in recovering OTUs from simulated and real-world metagenomic data executed on a mobile device.
• Showcased the feasibility of real-time, in-field metagenomic analysis using portable sequencers and mobile computing.
• Highlighted the benefits of incorporating feedback loops within the DNA analysis pipeline for improved performance.

Conclusions:

• The developed memory-efficient approach enables accurate OTU identification on mobile devices, facilitating field-based metagenomic studies.
• The integration of streaming algorithms and mobile-tailored pipelines overcomes limitations of traditional batch-processing tools.
• Real-time metagenomic analysis on portable platforms is achievable, with potential for further optimization through pipeline feedback loops.