Taxon composition and biodiversity analyses are known powerful parameters for environmental site status and environment diagnosis. Many ecological studies assess taxon composition through traditional species identification and use bioindicator species to evaluate environmental conditions. The recent breakthrough in bulk sample sequencing combined with DNA barcoding has created a new era for environmental monitoring. Metabarcoding approaches are more robust in studying alpha, and beta diversity compare to the DNA barcoding and the conventional method of species identification, particularly for rare and cryptic species. Here we built upon ecological studies of bioindicator species and transferred the traditionally named taxa to DNA-based approaches. We developed a small customized DNA database for biodiversity assessment and taxonomic identification of environmental DNA samples using high-throughput amplicon sequences. It contains macroinvertebrate species that are known as indicators of specific environmental conditions. By implementing this small database into the KRAKEN algorithm for the first time, we were able to assess environmental biodiversity compared to other popular methods of taxonomic classification, especially in polluted environments where the taxonomic composition globally change by the presence of anthropogenic drivers. Our method is incredibly faster, and it requires significantly less computational power in contrast to common homology-based techniques. To evaluate our approach, we have also studied the importance of database’s size and the depth of sequencing in taxonomic classification of high-throughput DNA sequences. / Thesis / Master of Science (MSc) / We developed a small customized DNA database for biodiversity assessment and taxonomic identification of environmental DNA samples using high-throughput amplicon sequences. It contains macroinvertebrate species that are known as indicators of specific environmental conditions. By implementing this small database into the KRAKEN algorithm for the first time, we were able to assess environmental biodiversity compared to other popular methods of taxonomic classification, especially in polluted environments where the taxonomic composition globally change by the presence of anthropogenic drivers. Our method is incredibly faster, and it requires significantly less computational power in contrast to common homology-based techniques.