NanoHIVSeq: A Long-Read Bioinformatics Pipeline for High-Throughput Processing of HIV Env Sequences

High-throughput sequencing of the HIV-1 envelope (Env) gene from viral quasispecies is essential for epidemiology, virus-antibody coevolution studies, and evaluating therapeutics, but the conventional single-genome amplification (SGA) coupled with Sanger sequencing is labor-intensive and low-throughput. Oxford Nanopore Technology (ONT) offers long-read sequencing advantages, but high error rates (1-7%) poses a challenge in identifying biological variants from sequencing artifacts. Without unique molecular identifiers (UMIs), which lose DNA template and add complexity in library preparation, here we introduce NanoHIVSeq, a UMI-free and reference-free bioinformatics pipeline that processes ONT data from bulk Env PCR amplicons through multistep clustering, consensus polishing, indel correction, denoising, and genotyping to recover functional full-length Env variants. By leveraging advanced ONT duplex sequencing technology, NanoHIVSeq was assessed using plasmid env and bulk HIV reservoir datasets, demonstrating high robustness, recovery rate, reproducibility, and accuracy (>99.9% or >Q30) comparable to UMI approaches. Our findings indicated that NanoHIVSeq allows flexible and simplified ONT library preparation for reproducible and efficient Env sequencing especially for large cohorts.