Fig 1.
TE and HTT distributions in relation with host phylogenetic diversity.
The tree is a chronogram based on Bailly-Bechet et al. [55]. Starting from the center, the first layer of data indicates the distribution of TEs (number of families per host species). The second layer indicates the estimated number of HTTs per species (mean number across 1000 plausible HTT scenarios). Boxes 1 and 2 represent arthropod orders with small sample size. Box 1 (from top to bottom): Pscocoptera (n = 10), Thysanoptera (n = 5), Orthoptera (n = 3), Mantodea (n = 1), Blatodea (n = 5), Isoptera (n = 1). Box 2 (from top to bottom): Amphipoda (n = 2), Scolopendromorpha (n = 1), Sarcoptiformes (n = 2), Ixodida (n = 1).
Table 1.
Minimal number of HTTs across superfamilies, in relation with their respective molecular diversities as measured by the total number of families.
Fig 2.
A graphical description of our HTT detection procedure.
Panel A: a general overview. We first group into families highly similar TEs (>80% nucleotidic identity) occurring in different species. Following a previously proposed formalism [43], we then establish networks of species sharing horizontally transferred TEs, based on dS comparisons of TEs versus housekeeping genes. Finally, we use the host phylogeny to compute the minimal number of HTTs, assuming that any cases of TE shared through HT between two sister clades can be explained by a single HT event. Panel B: details of the second step. The identification of TE families shared through HT first implied computing a tree including all species from our sample (labeled in black) in addition to arthropod species for which a complete genome was available (labeled in red), using sequences of the mitochondrial gene CO1 with imposed topology for deep nodes [55] (S1 Fig). The Most Recent Common Ancestor (MRCA) of Species X and Y (red dot) is the same as that of species 2 and 3. In other words, the divergence time between species X and Y is the same as that between species 2 and 3. We can thus use the distribution of dS values observed between housekeeping genes in genomes X and Y (panel B) to test the null hypothesis that TEs shared between species 2 and 3 have been inherited vertically. On the contrary, we cannot test this hypothesis for TEs shared between species 1 and 2. The MRCA of species 2 and 4 is older than the MRCA of species X and Y; we can thus use the X-Y dS distribution to conservatively test the hypothesis that TEs between species 2 and 4 have been inherited vertically. In this toy example, species 1 and 3 and species 1 and 4 share TEs that have less than 5% chance to result from vertical inheritance. We thus infer two cases of TEs shared through HT.
Fig 3.
A comparison between the observed and simulated number of species involved in HTT in each arthropod order, and for each superfamily.
Horizontal segments link the 2.5% and 97.5% quantiles of simulated values.