LiME

LiME performs alignment-free, assembly-free metagenomic classification of Next Generation Sequencing (NGS) reads by using an extended Burrows-Wheeler transform (eBWT) to measure sequence similarity and identify microorganisms in environmental and microbiome samples.

Key Features:

• eBWT-based similarity measurement: Uses an extended Burrows-Wheeler transform (eBWT) in a combinatorial approach to compute sequence similarity.
• Alignment-free and assembly-free classification: Classifies reads without relying on alignment or assembly steps.
• Low memory usage: Achieves minimal internal memory usage by sequentially scanning required data structures.
• Parallel processing support: Supports parallel processing across multiple processors or cores.
• Validation on NGS data: Tested on Next Generation Sequencing (NGS) data from simulated metagenomes and a real metagenome from the Human Microbiome Project.
• High precision and specificity: Assigned over 99.9% of positive control reads correctly and produced less than 0.01% misclassification in negative controls.
• Comparable sensitivity: Reported sensitivity comparable to the taxonomic classifier MagicBlast.

Scientific Applications:

• Metagenomic taxonomic profiling: Identification of microorganisms in environmental samples for studies in agriculture and ecology.
• Microbiome analysis: Characterization of human microbiome composition using datasets such as the Human Microbiome Project.
• Benchmarking and validation: Evaluation of classifier precision, specificity, and sensitivity using simulated metagenomes and control experiments.
• Large-scale NGS analysis: Resource-efficient analysis of large collections of Next Generation Sequencing (NGS) metagenomic datasets.

Methodology:

Applies a combinatorial extended Burrows-Wheeler transform (eBWT) approach to measure sequence similarity via sequential scanning of required data structures in an alignment-free and assembly-free framework and supports parallel processing across multiple processors/cores.