Mapping genetic determinants of viral traits with FST and quantitative trait locus (QTL) approaches

This research proves that FST and QTL can be used in virology

Read the full article on ScienceDirect


To identify the determinism of a trait, geneticists studying fungi, plants or animals generally rely on standard approaches (QTL mapping) based on progenies derived from sexual crossings. Because viruses multiply clonally, virologists address similar questions by measuring the consequences of introducing one or few mutations in the genome or of swapping genome fragments between two viral isolates, which is often tedious. Using 47 mosaic genomes randomly generated from two plant viruses with contrasted within-host accumulations, we showed that QTL mapping approaches used for eukaryotes could be exploited to identify a locus governing within-host viral accumulation. Because the same locus had been previously identified by classical virological approaches, this study provides a proof of principle that QTL mapping approaches can be used to identify the genetic determinants of viral traits. With the growing availability of high-throughput methods to sequence and phenotype many recombinant viruses generated in vitro or in vivo, such methods should become popular among virologists.


When we analyzed the effect of recombination on virus accumulation in a previous study (Vuillaume et al., 2011), we noticed that the phenotype of each progeny genome was significantly different from one parent but not from the other, which is characteristic of a single-locus dominant genetic determinism. We thus thought that classical QTL and FST approaches might be used to locate the corresponding locus.


According to our plant geneticist colleagues, the low number of clones on which phenotypic data were available did not a priori lend itself to the use of QLT analysis. In addition no one in the team was trained in the use of QTL software. Thus, the project remained on standby until J. Doumayrou decided to take up the challenge and use different available algorithms for QTL analysis. To do so, she had to familiarize herself to the principle and use of the different algorithms and adapt the parameters to virus genomes.


We became really excited when we saw that the data could be encoded in a way which was compatible with the available QTL mapping software. The fact that both FST and QTL mapping approaches pointed to the same genomic region, which in addition was consistent with previous virological knowledge, was also reassuring about the ability of the software to deal properly with viral data. Finally, we enjoyed the robustness showed by the approach when applied to another viral trait with a less obvious determinism, as well as the good feedback and advices received from geneticists.

figure Virology highlight QTL-V2

Figure legend

(A) Symptoms induced by Tomato yellow leaf curl virus (Tyx) and Tomato leaf curl Mayotte virus (Tox) and (B) their respective viral accumulation on tomato plants at 21 dpi.

(C) Schematic representation of recombinant viral genomes generated by L-DNA-shuffling with the parental genomes (Tyx and Tox). Each raw represents the genome of a recombinant clone. (D) Locus-by-locus FST between the genomes of 2 different accumulation groups and (E) QTL mapping using the single-marker regression QTL method identify the same viral region involved in within-host accumulation.

Introducing the author


Juliette Doumayrou obtained her PhD from the University of Montpellier in plant virus interaction.  During her PhD studies she worked with Yannis Michalakis on the evolution of virulence of Cauliflower mosaic virus. Currently, Juliette is a postdoctoral researcher exploring different aspects of virology as virulence, emergence, host-pathogen interactions and vector-borne transmission.

About the research

Mapping genetic determinants of viral traits with FST and quantitative trait locus (QTL) approaches

Virology, Volume 484, October 2015, Pages 346–353

Juliette Doumayrou, Gaël Thébaud, Florence Vuillaume, Michel Peterschmitt, Cica Urbino


Leave a Reply

Your email address will not be published. Required fields are marked *