Detecting and phasing minor single-nucleotide variants from long-read sequencing data

Cellular genetic heterogeneity is common in many biological conditions including cancer, microbiome, and co-infection of multiple pathogens. Detecting and phasing minor variants play an instrumental role in deciphering cellular genetic heterogeneity, but they are still difficult tasks because of technological limitations. Recently, long-read sequencing technologies, including those by Pacific Biosciences and Oxford Nanopore, provide an opportunity to tackle these challenges. However, high error rates make it difficult to take full advantage of these technologies. To fill this gap, we introduce iGDA, an open-source tool that can accurately detect and phase minor single-nucleotide variants (SNVs), whose frequencies are as low as 0.2%, from raw long-read sequencing data. We also demonstrate that iGDA can accurately reconstruct haplotypes in closely related strains of the same species (divergence ≥0.011%) from long-read metagenomic data.