Gretel

Gretel reconstructs haplotypes from metagenomic sequencing data to recover all gene isoforms across microbial community samples for population-level genetic variation analysis.

Key Features:

• Haplotype Recovery: Recovers genomic sequences of all isoforms of a given gene across multiple organisms within a metagenomic sample.
• Probabilistic Reconstruction: Parses read alignments into a Hansel matrix and uses evidence from SNP pairs observed on the same reads to probabilistically infer haplotypes.
• Pairwise SNV Co-occurrence Matrix (Hansel): Represents co-occurrence evidence between variant sites using the Hansel pairwise SNV co-occurrence matrix.
• Greedy Graph Traversal Algorithm: Uses a greedy graph traversal algorithm to navigate the Hansel-derived graph and reconstruct haplotype paths.
• Implementation: Implemented as a Python-based bioinformatics package.

Scientific Applications:

• Microbial Ecology and Evolution: Enables analysis of genetic variation within microbial communities to inform ecological roles and evolutionary dynamics.
• Medicine and Biotechnology: Supports metagenomic haplotyping applications relevant to medical research and biotechnology, including development of new antibiotics or probiotics.

Methodology:

Builds on single-individual haplotyping (SIH; Lancia et al., 2001) and extends this formalism to metagenomic individual haplotyping (MIH); parses read alignments into the Hansel pairwise SNV co-occurrence matrix, uses SNP pairs observed on the same reads for probabilistic reconstruction, and applies a greedy graph traversal algorithm to recover haplotypes.