Comparative mitogenomic analysis of the Drosophila buzzatii cluster.

JUAN HURTADO; NICOLÁS NAHUEL MOREYRA; ESTEBAN HASSON; JULIAN MENSCH

Los Álamos

Conferencia; Evolution and Ecology: The Cactophilic Drosophila Model System; 2018

CINVESTAV & University of San Diego

MtDNA has became a popular marker in evolutionary biology and population genetics given its high mutation rate respect to nuclear genetic markers, which allow to infer phylogenetics relationships between species and populations. Also, the lack of recombination, linear ancestry (inheritance) and high copy number frequently makes mitochondrial DNA the molecular marker of choice (Ladoukakis et al. 2017).The aim of the project is to design a workflow to reconstruct and obtain, for the first time, the mitochondrial complete genomes assemblies of four species of the Drosophila buzzatii cluster directly from genomic and transcriptomic sequencing data. Subsequently, it were studied the phylogenetic relationships among species and compared the results with the small set of mitochondrial markers previously reported (Oliveira et al. 2012).For this project, it were employed Paired Ends Illumina reads from bothgenomic and transcriptomic sequences. All the process can be summarizedby the next five steps:(1)Mapping genomic and transcriptomic reads to D. mojavensis mitochondrialgenome with bowtie2.(2)Extracting only confident mapped reads with Samtools and join to createone dataset per each species.(3)Performing an assembly with Mira assembler (B. Chevreux et al. 2014),using D. mojavensis as reference. At this stage, it is important to trydifferent genome coverage and set the Mira parameters.(4)Running MITObim (Hahn et al. 2013) with the complete set of reads oneach species to re-assembly the uncompleted regions of the resultingassembly in step 3. This software manage the genome coverage and set theoptimal value for each dataset.(5)Carrying out the genome annotation using the Mitos WebServer (Bernt etal. 2013) and the gene annotations of D. mojavensis.The resulting assemblies allow us to perform a phylogenetic analysis with the complete mitochondrial genomes of D. buzzatii, D. koepferae, D. antonietae and D. borborema, using D. mojavensis as outgroup. In addition, in order to validate the annotated genes, we compared the mitochondrial genes previously reported by Oliveira et al. (2012) (nad2, cox1, cox2 and lsRNA) with the respective assemblies obtained in this project.The statistics for the assemblies are shown in Table 1. At least the 90% the total sequence of each mitogenome was assembled and all annotated genes on each corresponding assembly were in the correct order and sense. Phylogenetic tree using the complete mitochondrial genomes are similar with those obtained by morphological characters and nuclear sequences, being D. buzzatii the most divergent species of the cluster (Figure 1). Importantly, our annotated genes perfectly matched with those obtained by Oliveira et al. (2012) and formed groups in each species (Figure 2) although we need to employ the whole mitogenomes to reconstruct a high-confidence phylogeny (Figure 1).The unknown regions (N´s) must be analyzed by other methods given the variability in the total length. It should be noted that the less performance assembly (D. buzzatii) was only made with transcriptomic sequencing data, highlighting the relevance of having genomic information, probably because of its greater enrichment in mitochondrial sequencing reads. As a future perspective, it is planned to sequence the unrevealed regions. This action will allow to evaluate consistency of the assemblies and filling the unknown bases.