swga2_0

swga2_0 implements an optimized, machine-learning-guided primer design pipeline for selective whole genome amplification (SWGA) to amplify microbial genomes from mixed samples for next-generation sequencing.

Key Features:

• Machine Learning Integration: Incorporates active learning and machine learning techniques to evaluate the efficacy of individual primers and primer sets.
• Optimization and Parallelization: Optimizes primer-set search and evaluation strategies by parallelizing each stage to reduce runtime.
• Empirical Data Utilization: Leverages empirical amplification data to identify primer and primer-set characteristics that improve performance.

Scientific Applications:

• Microbial population genomics: Enables generation of sufficient microbial genomic DNA for population-level genomic studies at fine spatial and temporal scales.
• Complex-sample target recovery: Facilitates selective amplification of target microbial genomes from samples with predominant non-target DNA (e.g., human DNA).
• Evolutionary and mechanistic studies: Supports investigations of evolutionary and mechanistic processes by improving access to target microbial genomes for sequencing.
• Target organism demonstration: Applicable to organisms such as Prevotella melaninogenica recovered from mixed clinical samples.

Methodology:

Designs primer sets for SWGA using active learning and machine learning to evaluate primers and primer sets, optimizes primer-set search and evaluation via parallelization, and uses empirical amplification data to identify effective primer characteristics.