To date, different strategies of whole-genome sequencing (WGS) have been developed in order to understand the genome structure and functions. However, the analysis of genomic sequences obtained from natural populations is challenging and the biological interpretation of sequencing data remains the main issue. The MinION device developed by Oxford Nanopore Technologies (ONT) is able to generate long reads with minimal costs and time requirements. These valuable assets qualify it as a suitable method for performing WGS, especially in small laboratories. The long reads resulted using this sequencing approach can cover large structural variants and repetitive sequences commonly present in the genomes of eukaryotes. Using MinION, we performed two WGS assessments of a Romanian local strain of Drosophila melanogaster, referred to as Horezu_LaPeri (Horezu). In total, 1,317,857 reads with a size of 8.9 gigabytes (Gb) were generated. Canu and Flye de novo assembly tools were employed to obtain four distinct assemblies with both unfiltered and filtered reads, achieving maximum reference genome coverages of 94.8% (Canu) and 91.4% (Flye). In order to test the quality of these assemblies, we performed a two-step evaluation. Firstly, we considered the BUSCO scores and inquired for a supplemental set of genes using BLAST. Subsequently, we appraised the total content of natural transposons (NTs) relative to the reference genome (ISO1 strain) and mapped the mdg1 retroelement as a resolution assayer. Our results reveal that filtered data provide only slightly enhanced results when considering genes identification, but the use of unfiltered data had a consistent positive impact on the global evaluation of the NTs content. Our comparative studies also revealed differences between Flye and Canu assemblies regarding the annotation of unique versus repetitive genomic features. In our hands, Flye proved to be moderately better for gene identification, while Canu clearly outperformed Flye for NTs analysis. Data concerning the NTs content were compared to those obtained with ONT for the D. melanogaster ISO1 strain, revealing that our strategy conducted better results. Additionally, the parameters of our ONT reads and assemblies are similar to those reported for ONT experiments performed on various model organisms, revealing that our assembly data are appropriate for a proficient annotation of the Horezu genome.