Libra: Comparative Metagenomics With MapReduce

Libra is an algorithm for all-vs-all genome sequencing tool to identify problematic microbes. It handles large data sets with speed, accuracy and efficiency.

Background:

Most bioinformatics methods need a reference source of known genomes to ensure accurate measurements of depth and variance. Sequencing and comparing wholly unknown genomes with no reference (all-vs-all) can be unwieldy both in terms of computation and storage, making microbes in complicated samples hard to identify. Libra is constructed on a big data framework, which makes it tailored to scale up for large sequences and efficiently store them.

Applications:

• All-vs-all sequencing, specifically microbe identification in complex and varied samples
• Drug discovery, taxonomy and pathology
• Genomic sequencing and information
• Libra can be a catch all for accurate microbe identification for any purpose, as it does not use probability methods which need reference data

Advantages:

• Does not use probability methods
• Reference-free
• Easy to use
• A catch-all for accurate microbe identification for any purpose
• Lower overhead costs
• Useful for many purposes involving microbe analysis even outside of pathology and taxonomy
• Easily scalable