Recombination is one of the determinants of genetic diversity in the foot-and-mouth disease virus (FMDV). FMDV sequences have a mosaic structure caused by extensive intra- and inter-serotype recombination, with the exception of the capsid-encoding region. While these genome-wide patterns of broad-scale recombination are well studied, not much is known about the patterns of recombination that may exist within infected hosts. In addition, detection of recombination among viruses evolving at the within-host level is challenging due to the similarity of the sequences and the limitations in differentiating recombination from point mutations. Here, we present the first analysis of recombination events between closely related FMDV sequences occurring within buffalo hosts. The detection of these events was made possible by the occurrence of co-infection of two viral swarms with about 1% nucleotide divergence. We found more than 15 recombination events, unequally distributed across eight samples from different animals. The distribution of these events along the FMDV genome was neither uniform nor related to the phylogenetic distribution of recombination breakpoints, suggesting a mismatch between within-host evolutionary pressures and long-term selection for infectivity and transmissibility.

Foot-and-mouth disease virus (FMDV) causes a highly contagious disease of cloven-hooved animals that poses a constant burden on farmers in endemic regions and threatens the livestock industries in disease-free countries. Despite the increased number of publicly available whole genome sequences, FMDV data are biased by the opportunistic nature of sampling. Since whole genomic sequences of Southern African Territories (SAT) are particularly underrepresented, this study sequenced 34 isolates from eastern and southern Africa. Phylogenetic analyses revealed two novel genotypes (that comprised 8/34 of these SAT isolates) which contained unusual 5′ untranslated and non-structural encoding regions. While recombination has occurred between these sequences, phylogeny violation analyses indicated that the high degree of sequence diversity for the novel SAT genotypes has not solely arisen from recombination events. Based on estimates of the timing of ancestral divergence, these data are interpreted as being representative of un-sampled FMDV isolates that have been subjected to geographical isolation within Africa by the effects of the Great African Rinderpest Pandemic (1887–1897), which caused a mass die-out of FMDV-susceptible hosts. These findings demonstrate that further sequencing of African FMDV isolates is likely to reveal more unusual genotypes and will allow for better understanding of natural variability and evolution of FMDV.

High-throughput sequencing such as those provided by Illumina are an efficient way to understand sequence variation within viral populations. However, challenges exist in distinguishing process-introduced error from biological variance, which significantly impacts our ability to identify sub-consensus single-nucleotide variants (SNVs). Here we have taken a systematic approach to evaluate laboratory and bioinformatic pipelines to accurately identify low-frequency SNVs in viral populations. Artificial DNA and RNA “populations” were created by introducing known SNVs at predetermined frequencies into template nucleic acid before being sequenced on an Illumina MiSeq platform. These were used to assess the effects of abundance and starting input material type, technical replicates, read length and quality, short-read aligner, and percentage frequency thresholds on the ability to accurately call variants. Analyses revealed that the abundance and type of input nucleic acid had the greatest impact on the accuracy of SNV calling as measured by a micro-averaged Matthews correlation coefficient score, with DNA and high RNA inputs (10⁷ copies) allowing for variants to be called at a 0.2% frequency. Reduced input RNA (10⁵ copies) required more technical replicates to maintain accuracy, while low RNA inputs (10³ copies) suffered from consensus-level errors. Base errors identified at specific motifs identified in all technical replicates were also identified which can be excluded to further increase SNV calling accuracy. These findings indicate that samples with low RNA inputs should be excluded for SNV calling and reinforce the importance of optimising the technical and bioinformatics steps in pipelines that are used to accurately identify sequence variants.