Quality filtered data

The Neotropics harbors the largest species richness of the planet; however, even in well-studied groups, there are potentially hundreds of species that lack a formal description, and likewise, many already described taxa have challenging identifications. Specifically in small mammals, complex morphological diagnoses have been facilitated by the use of molecular data, particularly from mitochondrial sequences, to obtain accurate species identifications. Obtaining mitochondrial markers implies the use of PCR and specific primers, which are largely absent for non-model organisms. Oxford Nanopore Technologies (ONT) is a new alternative for sequencing the entire mitochondrial genome without the need for specific primers. Nevertheless, only a limited number of studies have employed exclusively ONT long-reads to assemble mitochondrial genomes, and no study has yet evaluated the usefulness of such reads in multiple non-model organisms. Hence, in this study, we explore the use of ONT long reads as a tool for the reconstruction of mitochondrial genomes of small mammals and a potential method for real-time species identification.We implemented fieldwork to collect small mammals, including rodents, bats, and marsupials, in five localities in the northern extreme of the Cordillera Central of Colombia. For DNA extraction six millimeters were removed from the distal section of the tail of terrestrial mammals (rodents, opossums), and three patagium (wing membrane) samples were taken from bats. DNA samples were sequenced using the MinION device and Flongle flow cells. Shotgun-sequenced data was used to reconstruct the mitochondrial genome of all the samples. In parallel, using a customized computational pipeline, species-level identifications were obtained based on sequencing raw reads (Whole Genome Sequencing or mitochondrial-filtered reads). ONT-based identifications were corroborated using traditional morphological characters and phylogenetic analysis. This dataset corresponds to the raw fastq sequence files with reads above Q7. Each file corresponds to one of the 21 individuals for which we obtained sequencing data.