*4.4. Phylogenetic Reconstruction*

Sequence alignment for all published and available Lentibulariaceae cpDNA genomes (Species and accession number, respectively: *Utricularia foliosa*, KY025562; *U. gibba*, KC997777; *U. reniformis*, KT336489; *U. macrorhiza*, HG803177; *Pinguicula ehlersiae,* HG803178, *Genlisea aurea*, MF593121; *G. violacea*, MF593126; *G. tuberosa*, NC\_037082; *G. filiformis*, MF593122; *G. pygmaea*, MF593123; *G. repens*, MF593124; *G. margaretae*, HG530134) were performed with online MAFFT v.7 [67] with default parameters. The phylogenetic tree reconstruction was performed using the Bayesian inference (BI) and maximum likelihood (ML) approaches under the best-of-fit model GTR+G+I, in accordance with Akaike Information Criterion (AIC), assessed by Mr.ModelTest v.2.4 [68]. For BI, set for the substitution model accordingly, we employed Mr.Bayes v.3.2.6 software with 5 <sup>×</sup> 107 generations sampled for each 1000 generations using two runs and four chains, until the average standard deviation of split frequencies became less than 0.01, beginning with random trees. The initial trees were discarded after reaching stationary (~25%). For ML we used the software RAxML v.8.2.10 [69] with 1000 bootstrap pseudoreplicates. The analyses were performed using implementations within CIPRES Science Gateway v.3.3 (https://www.phylo.org/) and cladograms were edited using TreeGraph v.2.15 beta [70]. The species *Tectona grandis* (NC020098), *Sesamum indicum* (NC016433), and *Tanaecium tetragonolobum* (NC027955) were used as outgroup.
