HAPHPIPE: Haplotype Reconstruction and Phylodynamics for Deep Sequencing of Intrahost Viral Populations

Abstract

Deep sequencing of viral populations using next-generation sequencing (NGS) offers opportunities to understand and investigate evolution, transmission dynamics, and population genetics. Currently, the standard practice for processing NGS data to study viral populations is to summarize all the observed sequences from a sample as a single consensus sequence, thus discarding valuable information about the intrahost viral molecular epidemiology. Furthermore, existing analytical pipelines may only analyze genomic regions involved in drug resistance, thus are not suited for full viral genome analysis. Here, we present HAPHPIPE, a HAplotype and PHylodynamics PIPEline for genome-wide assembly of viral consensus sequences and haplotypes. The HAPHPIPE protocol includes modules for quality trimming, error correction, de novo assembly, alignment, and haplotype reconstruction. The resulting consensus sequences, haplotypes, and alignments can be further analyzed using a variety of phylogenetic and population genetic software. HAPHPIPE is designed to provide users with a single pipeline to rapidly analyze sequences from viral populations generated from NGS platforms and provide quality output properly formatted for downstream evolutionary analyses.

Incorporating Within-Host Diversity in Phylogenetic Analyses for Detecting Clusters of New HIV Diagnoses

“…The other regions have ranges of 0.2–1.9% (mean 0.9%) for prrt, 0.1–2.0% (mean 0.9%) for int , and 0.2–2.6% (mean 1.2%) for wgs. Such within-host estimations are not feasible with conventional consensus Sanger or NGS approaches, although methods such as phyloscanner and HAPHPIPE that utilize deeper NGS sequencing to build phylogenies with multiple tips per sample, are able to also quantify within-host diversity ( Wymant et al, 2018 ; Bendall et al, 2021 ).…”

“…Two other studies introduced methods to use deeply-sequenced HIV data without assuming a consensus, for a different but related epidemiological goal of estimating transmission directionality and identifying multiple infections ( Skums et al, 2018 ; Wymant et al, 2018 ). Methods also exist that combine haplotype estimation from deeply-sequenced NGS data and phylogenetics ( Bendall et al, 2021 ) as a way to incorporate within-host diversity, but available haplotyping methods have a high computational cost and results are often not sufficiently accurate for cluster analysis ( Wymant et al, 2018 ). Additionally, all aforementioned methods that do not rely on a consensus incorporate within-host diversity by including multiple sequences or tips per sample, which presents difficulties with summarizing or collapsing the resulting phylogenetic tree in order to identify transmission clusters and measure cluster certainty.…”

V-pipe 3.0: a sustainable pipeline for within-sample viral genetic diversity estimation

“…We compare V-pipe 3.0 to other relevant viral bioinformatics pipelines for within-sample diversity estimation, focusing on functionalities and sustainability (Table 2). The compared pipelines include nf-core/viralrecon [10], HAPHPIPE [11] and ViralFlow [9]. These pipelines are all open source, actively maintained, and provide within-sample diversity estimates for Illumina sequencing reads.…”

“…Typically they combine tools for quality control, sequence alignment, consensus sequence assembly, diversity estimation, and result visualization. Various workflows have been proposed which try to accomplish these goals including V-pipe [8], ViralFlow [9], nf-core/viralrecon [10] and HAPHPIPE [11]. The adaptability of these workflows becomes crucial as different types of viruses require tailored analysis approaches.…”

Establishment and application of a method of tagged-amplicon deep sequencing for low-abundance drug resistance in HIV-1

“…To assess minor drug resistance mutations of HIV in vivo , single genome sequencing (SGS; McKinnon et al, 2011 ; Wang et al, 2014 ) and allele-specific real-time PCR (ASPCR; Hauser et al, 2015 ) were developed, but they cannot be applied to large-scale samples detection due to time-consuming and labor-intensive. In recent years, NGS is becoming a more common sequencing method that is widely used in the current genetic detection field ( Gibson et al, 2020 ; Bendall et al, 2021 ; Parikh et al, 2021 ). NGS can detect low-abundance drug resistance mutations by sequencing the HIV-1 quasispecies ( Ji et al, 2020 ).…”