Comparison of Next-Generation Sequencing and Traditional Melissopalynological Methods for Geographically Labeled Anzer Honey

Comparison of Next-Generation Sequencing and Traditional Melissopalynological Methods for Geographically Labeled Anzer Honey | Authorea try { document.documentElement.classList.add('js'); } catch (e) { } var _gaq = _gaq || []; _gaq.push(['_setAccount', 'G-8VDV14Y67G']); _gaq.push(['_trackPageview']); (function() { var ga = document.createElement('script'); ga.type = 'text/javascript'; ga.async = true; ga.src = ('https:' == document.location.protocol ? 'https://ssl' : 'http://www') + '.google-analytics.com/ga.js'; var s = document.getElementsByTagName('script')[0]; s.parentNode.insertBefore(ga, s); })(); Skip to main content Preprints Collections Wiley Open Research IET Open Research Ecological Society of Japan All Collections About About Authorea FAQs Contact Us Quick Search anywhere Search for preprint articles, keywords, etc. Search Search ADVANCED SEARCH SCROLL This is a preprint and has not been peer reviewed. Data may be preliminary. 28 February 2025 V1 Latest version Share on Comparison of Next-Generation Sequencing and Traditional Melissopalynological Methods for Geographically Labeled Anzer Honey Authors : Zeynep Türker 0009-0004-8666-7793 [email protected] , Kamil Coskuncelebi 0000-0001-6432-9807 , Murat Güzel , and Serdar Makbul Authors Info & Affiliations https://doi.org/10.22541/au.174074244.46911923/v1 355 views 155 downloads Contents Abstract Introduction Material and methods Reference database Data analysis Statistics NGS findings Comparison of melissopalynological and NGS data References Information & Authors Metrics & Citations View Options References Figures Tables Media Share Abstract This study utilized next-generation sequencing (NGS) of nrDNA ITS regions (ITS1 and ITS2) for the first time to analyze three honey samples from Anzer (Ballıköy), Rize province, Türkiye. The NGS results were evaluated alongside melissopalynological data. Pollen grains were first isolated and identified microscopically, and DNA was extracted from the honey samples for NGS analysis. ITS1 and ITS2 regions were sequenced using Illumina MiSeq, and results were compared with a custom reference library. NGS produced 310,745 paired-end reads for ITS1 and 39,835 reads for ITS2. Of these, 75.2% of ITS1 reads and 68.4% of ITS2 reads were identified to at least the family level. NGS analysis detected 27 plant families and 54 taxa, a 37% increase in taxa detection compared to melissopalynology, which identified 19 families and 34 taxa. Both approaches consistently identified dominant floral components, with NGS providing greater species-level resolution. Spearman’s correlation revealed a moderate linear relation between the two methodologies for two of the three samples. However, the Shannon-Wiener and Pielou indices were lower in metabarcoding than in melissopalynology due to the uneven distribution of read counts for some species. The R-coefficient results of all the families for the three samples showed over or underrepresentation except for Caryophyllaceae (honey sample ZT2 = 0.85) and Asteraceae (honey sample ZT3 = 0.93). While to date, melissopalynolgy has been the prime identification method for determining the geographical origin of honey, this study, for the first time, presents a comprehensive and reliable metabarcoding data for Anzer honey identification. Introduction Nowadays, the excessive interest in natural foods with high therapeutic and nutritional properties has led to an increase in the demand for natural products such as honey, which has been used for different medicinal purposes since ancient times (Eteraf-Oskouei & Najafi, 2013). In addition to the important role of honey in traditional nutrition, scientists also accept honey as an effective new medicine for many kinds of diseases. This situation leads to the imitation of many local honeys such as Anzer honey, which stands out with its biological properties. This is because there is an increasing need to safeguard consumers and promote fair competition among producers, particularly concerning identifying the geographic origin of honey. Anzer honey is produced by beekeepers in the Anzer (Ballıköy) village (Rize/Türkiye) known for its rich plant diversity with over 500 flowering plants, around 100 of which are endemic and special vegetations (Ulusoy & Kolaylı, 2014; Terzioğlu, 1994). Anzer honey, categorized as multifloral honey (Malkoç et al., 2019) have a light hue and a unique, aromatic flavor. The honey is highly valued since it is considered a functional food with numerous health benefits (Çil, 2023; Hepsağ, 2019; Malkoç et al., 2019; Ulusoy & Kolaylı, 2014), owing to the rich flora of Anzer (Ballıköy). Anzer honey is produced in limited quantities by the beekeepers located in the Anzer village due to the special vegetation and climatic conditions (Rize İl Tarım ve Orman Müdürlüğü, 2019). The limited production of Anzer honey increases demand and makes it more expensive and prone to imitations. Anzer honey is one of the rare local honeys with a geographical indication based on traditional palynological data (Sorkun & Doğan, 1995); however, this traditional analysis has some drawbacks, including being time-consuming, requiring extensive expertise necessary to distinguish different pollen grains, and relying heavily on specialists with sufficient knowledge to distinguish pollen grains belonging to different plant taxa (Bell et al., 2016). Chemical analysis, even though used for determining the antioxidant and antimicrobial activities of Anzer honey (Ulusoy et al., 2010; Ulusoy & Kolaylı, 2014) are not supplied sufficient data for its botanical and geographical origin. The chemical composition of honeys with the same botanical origin varies greatly depending on factors such as phytochemicals, soil characteristics, flower age, etc (Balkanska et al., 2020). Therefore, chemical analysis methods are insufficient to determine the botanical origin of honey with certainty. In recent years, there has been increasing interest in accurately determining the geographical origins of different honeys. Recent advances in DNA sequencing methods have drawn a lot of interest because of their speed, reliability, and capacity to discriminate plant taxa at a specific or higher level without the need for expertise in taxonomy (Bruni et al., 2015; Schnell et al., 2010). Various DNA-based techniques have been employed to identify plant species in honey. These include real-time PCR (Laube et al., 2010), PCR followed by Sanger sequencing (Wilson et al., 2010), and PCR amplification with cloning and sequencing (Bruni et al., 2015; Olivieri et al., 2012; Schnell et al., 2010) and pollen grains (Galimberti et al., 2014), each with specific advantages and limitations. Hawkins et al., (2015) used the Rbc L DNA marker to compare melissopalynology with Illumina and 454 pyrosequencing techniques for 9 honey samples from Wales England. Comparative studies by Keller et al., (2015) and Kraaijeveld et al., (2015), which compared classical methods with NGS using ITS2 and trnL, respectively, indicated that NGS offers a viable alternative for identification. Although there are several studies on Anzer honey regarding its electrical conductivity (Hotaman, 2015), phenolic, antioxidant, antimicrobial (Hepsağ, 2019; Ulusoy et al., 2010; Ulusoy & Kolaylı, 2014), and palynological properties (Güner et al., 1987; Sorkun & Doğan, 1995), there are a limited number of studies regarding the molecular characterization of Anzer honey. Research conducted by Özkök et al., (2023) comparing melissopalynology with Next Generation Sequencing (NGS) data, utilizing rbc L and trn H- psb A markers across 74 diverse honey samples from different parts of Türkiye, represents a promising advancement in this direction. The current study aims to assess the feasibility of using next-generation sequencing as an alternate characterization method to determine the floral composition of Anzer honey and compare the results with those obtained by the traditional method (melissopalynology). The data from this study can act as a scientific resource for the prevention of honey adulteration for the geographically labeled Anzer honey (Rize/Türkiye). Material and methods Honey sampling Honey samples (ZT1, ZT2, ZT3) used in the study were obtained from domestic beekeepers located in the Anzer village (Rize, Türkiye). The first two honey samples (ZT1, ZT2) were a mixture of honey from different producers’ hives, while the third sample (ZT3) belonged to a single producer’s hives. All samples were stored in glass jars at room temperature in the Botany Laboratory of the Department of Biology, Karadeniz Technical University before palynological and DNA extraction studies. For palynological and DNA analyses, 10 g and 50 g of honey were used from the stock for each sample using a large sterile syringe, respectively. Counting total pollen grains 10 g honey was initially mixed with 20 mL of distilled water alongside a tablet containing 9666 Lycopodium spores followed by incubation at 45°C for 10-15 minutes. After dissolution, a few drops of basic fuchsin were added for staining, and the mixture was centrifuged at 3500 rpm for 45 minutes (Sorkun, 2008). The supernatant was discarded, and 0.1 mL of 50% glycerin was added to the pellet and mixed thoroughly. Then 0.01 mL of this mixture was taken and the final volume was adjusted to 1 mL with 50% glycerin. The mixture was finally transferred onto glass slides and covered by a lamella. These slides were then examined under a light microscope, scanning an 18x18 mm 2 area to quantify pollen grains, while Lycopodium spores were counted separately (Sorkun, 1985). The categorization of the honey samples based on the total number of pollen grains (TPN) was performed according to the guidelines of Sorkun, (1985) as in the following: low (>20 000), normal (20 000-100 000), rich (100 000-500 000) and very rich (500 000-1 000 000). Melissopalynology Melissopalynological studies were performed by modifying and optimizing Wodehouse (1935) method. Firstly 10 g of honey taken from each sample was properly mixed using a sterile glass rod prior to being transferred to a 20 mL centrifuge tube. Subsequently, 10 milliliters of distilled water was added to each tube. To ensure complete dissolution, the samples were agitated in a water bath at 45°C for 30-45 minutes. Afterward, the solution was then centrifuged at 3500 rpm for 45 minutes and the supernatant was discarded. The sediment remaining at the bottom of the tubes was impregnated with glycerin-gelatin containing basic fuchsin and then transferred to glass slides. To facilitate the dissolution of basic fuchsin-added glycerin-gelatin, the glass slides were heated at 30-40°C and covered with a coverslip at a 45° angle. Even though no specific reference slides were created for the study, pollen reference slides from the Anzer village, and published resources like the Palynological Database (PalDat) (PalDat, 2023), Pollen-Wiki (Pollen-Wiki, 2025) and the book Nectar Plants, Pollen, and Honey of Turkey (Sorkun, 2008) were utilized for identification purposes. Great care was taken to ensure the terminological consistency. An attempt was made to identify pollen to the lowest possible taxonomic level. However, in many instances, certain pollen types could only be classified to the genus or family level, and in some cases, identification was not possible, though all pollen grains were still included in the analysis. All slides were examined under a light microscope using 10x, 40x, and 100x magnifications. An average of around 130 pollen grains were observed on each slide. The distribution of the pollen grains according to taxonomic level (family, genera, and species) was determined as the percentages, and then honey samples were classified as monofloral or multifloral based on these findings (Louveaux et al., 1978). DNA extraction The pre-treatment of the samples was performed as described by Cheng et al., (2007) with small modifications. Total DNA extraction from each honey sample was performed by modifying and optimizing the traditional CTAB method (Utzeri et al., 2018). Initially, 50 g of honey samples were evenly distributed into four separate falcon tubes, each containing 12.5 g, followed by adding 45 mL of distilled water. After vortexing for 20 seconds and 1 hour incubation at 65◦C, the first centrifugation was carried out for 20 minutes at 14,000 g with subsequent removal of the supernatant. The resulting pellets were resuspended in 20 mL distilled water and subjected to a second centrifugation at 14,000 g for 10 minutes. After discarding the supernatant, the pellets from the four tubes (four tubes for each sample) were combined into a single clean tube to which 5 mL of distilled water was added. The combined pellets underwent a third centrifugation at 14,000 g for 5 minutes at room temperature. After discarding the supernatant, the precipitates were reconstituted in 1000 μL of distilled water and incubated overnight at 4°C in 2 mL tubes. The next day, the combined samples were air-dried for 30 minutes after another centrifugation at 14,000 g for 15 minutes. Subsequently, each tube was filled with at least four pieces of 0.3 cm sterile metal beads and sterile glass powder. They were then submerged in liquid nitrogen for fifteen minutes. After that, the samples were grounded for 5 minutes using a homogenizer at 50 Hz. Following these steps, the glass beads were taken out of the tubes and allowed to return to room temperature. The remaining pellets were then mixed with 500 μL of CTAB extraction buffer, containing 20 μl EDTA, 50 μl Tris, 140 μl NaCl, 100 μl CTAB, and 190 μl H 2 O. Additionally, 0.02 g of PVPP (Polyvinylpolypyrrolidone), 30 μl of proteinase K, 2.5 μl of RNase, and 2.5 μl of β-mercaptoethanol were then added to the mixture. The resulting mixture was homogenized using a pipette and then incubated at 65°C for 4 hours. The rest of the steps associated with the isolation were followed as described by Utzeri et al., (2018). After DNA isolation, its integrity was detected using 1% agarose gel electrophoresis. The purity, and concentration (A260/A280 values) of DNA were determined using the Nanodrop spectrophotometer (ThermoFisher Scientific, Canada). Amplification and sequencing DNA barcoding analysis was performed using the nrDNA ITS regions, specifically ITS1 and ITS2. The process involved two rounds of PCR. The first round amplified the ITS1 and ITS2 regions using universal primers with adaptor sequences (Table 1). Table 1. The universal ITS primers used in the study along with the added adaptors designated in bold. Adaptor+ITS1 ITS5N F TCGTCGGCAGCGTCAGATGTGTATAAGAGACAG GGAAGTAAAAGTCGTAACAAGG ITS2N R GTCTCGTGGGCTCGGAGATGTGTATAAGAGACAG GCTGCGTTCTTCATCGATGC Adaptor+ITS2 ITS3N F TCGTCGGCAGCGTCAGATGTGTATAAGAGACAG GCATCGATGAAGAACGCAGC ITS4N R GTCTCGTGGGCTCGGAGATGTGTATAAGAGACAG TCCTCCGCTTATTGATATGC The PCR master mix contained 3.5 μL MgCl 2 (2.5 mM), 10 μL 10× buffer, 20 μL dNTP (0.25 mM), 3 μL gelatin, 1 μmol/L forward and reverse primers, 0.3 μL Taq Phusion polymerase, and 9.7 μL distilled water, for a total reaction volume of 50 μL. The PCR conditions included an initial denaturation at 95 °C for 2.5 minutes, followed by 36 cycles of denaturation at 94 °C for 60 seconds, primer annealing at 55 °C for 30 seconds, and DNA synthesis at 72 °C for 540 seconds, with a final extension step at 72 °C for 10 minutes. PCR products were verified using agarose gel electrophoresis on 0.8% (w/v) agarose gels. Before proceeding to the second round of PCR, samples were purified using the AMPure XP Beads kit to remove DNA fragments and potential primer dimers. During the second round, Illumina sequencing adapters with i5 and i7 dual index barcodes were added to the amplicons (Figure 1). These index barcodes allowed for the simultaneous processing of multiple samples during sequencing. Additionally, the Illumina adapters contained sequences that facilitate binding to the flow cell surface, ensuring proper clustering and sequencing efficiency. To guarantee the results’ validity, ITS1 and ITS2 contained negative and positive PCR controls. The final concentrations of the PCR products were quantified using a Qubit 3.0 Fluorometer. Figure 1. DNA fragments with adapters attached to the ends. Reference database To improve taxonomic resolution and minimize erroneous reads, a local reference database of sequences was compiled from GenBank (Benson et al., 2015). While creating this reference database, specifically ITS data belonging to wild flowering plants distributed in Rize province, native and alien species, ornamental plants, horticultural and agricultural plants widely distributed in the Eastern Black Sea Region and throughout Turkey were used (Anonymous, 2014; Güner et al., 2012; Terzioğlu, 1994). A total of around 2000 sequences were cumulated. Due to the potential for incorrect sequences in GenBank data, each sequence was carefully verified, and any corrupted sequences were eliminated from the reference database. Data analysis Two sets of PCR products, each 251 bp long, were sequenced using paired-end Illumina MiSeq libraries with indexing. Sequencing yielded an initial output of 200,000 paired-end reads (251 bp per read) per sample. A quality control analysis of the raw FASTQ data was carried out with FastQC (version 0.11.9) using default parameters. For bioinformatics analyses, the Linux-based QIIME 2 platform was used. The raw data with attached adapters was trimmed and introduced into the QIIME 2 platform (Figure 2). Quality filtering, merging forward and reverse reads, chimerism detection and filtering, and noise reduction were performed with q2-dada2 to obtain Amplicon Sequence Variants (ASVs). Figure 2. Bioinformatic pipelines used for the NGS analysis. Statistics To compare the findings of melissopalynology and NGS, the data from both methods was first transformed into their relative abundances. Spearman’s correlation analysis was employed to ascertain the intensity and direction of the relationship between the data from the two methods. Furthermore, Shannon-Wiener (H’) and Pielou’s evenness indices (J’) were employed to assess the diversity and homogeneity of the taxa, respectively (Smart et al., 2017). The maximum diversity (H max ) of each sample was determined by H max = ln(S); where S is the number of species. The range of Pielou’s evenness varies from zero to 1, with 1 representing equal representation of all the taxa. For a given taxon, the R coefficient is the ratio of the percent mapped reads in NGS to the percent relative abundance of the pollen found in microscopy (Richardson, et al. 2015). According to metabarcoding and microscopy results for the families detected in the examined honey samples, R coefficients were calculated to determine which families were over or under-represented. not-yet-known not-yet-known not-yet-known unknown 1 Results 1.1 Melissopalynological findings The palynological analysis revealed that the honey samples ZT3 and ZT1 contained the highest (246,483) and the lowest (10,512) total pollen counts respectively while the ZT2 honey sample had 76,311 pollen grains (Figure 3). 1 Results Figure 3 . Total pollen number (TPN) of the three honey samples. Using the traditional melissopalynological method, approximately 130 pollen grains from various flowering plants belonging to different taxonomic levels, were identified in the three honey samples (Table 2). Most of the pollen grains (94-100%) were characterized at the specific, generic, or family level (Table 3). Of these, 3% were identified at the specific level, 60% at the generic level, and 37% at the family level. All three honey samples contained taxa represented by a single pollen grain or singleton. Sample ZT2 had the most singletons (9) compared to samples ZT1 and ZT3, which contained 1 and 5 singletons, respectively (Table 2). Table 2 . Melissopalynologic results of the three honey samples. Count % Count % Taxa Family % Genus % Species % Singleton % ZT1 129 94.2 8 5.8 15 6 40.0 8 53.3 1 6.7 1 6.7 ZT2 132 100.0 0 0 29 8 27.6 20 69.0 1 3.4 9 31.0 ZT3 127 99.2 1 0.8 14 6 42.9 8 57.1 0 0 5 35.7 Average 129.3 97.8 3 2.13 19.3 6.67 36.8 12 59.8 0.67 3.37 5 24.5 not-yet-known not-yet-known not-yet-known unknown Table 3 lists the pollen spectra of the examined samples at family, generic, and specific levels. Dominant (≥45%) taxa are shown as red, secondary (16–44%) as green, minor (3–15%) as yellow, and rare (<3%) as light yellow. Table 3. Melissopalynologic pollen spectrum of the three honey samples: Dominant (red), secondary (green), minor (dark yellow), trace (light yellow) Pollen count Pollen spectrum (%) Pollen count Pollen spectrum (%) Pollen count Pollen spectrum (%) Apiaceae Apiaceae 32 24.8 27 20.5 2 1.6 Asteraceae Asteraceae 43 33.3 7 5.3 2 1.6 Asteraceae Artemisia 3 2.3 Asteraceae Centaurea 5 3.8 Asteraceae Helianthus 1 0.8 Asteraceae Taraxacum 3 2.3 Boraginaceae Echium 2 1.5 Boraginaceae Myosotis 2 1.6 Brassicaceae Brassicaceae 5 3.9 22 16.7 Campanulaceae Campanula 12 9.3 2 1.5 Caryophyllaceae Caryophyllaceae 4 3.1 1 0.8 Caryophyllaceae Dianthus 2 1.5 Cistaceae Cistus 3 2.3 Cistaceae Helianthemum 4 3.1 3 2.3 6 4.7 Chenepodiaceae Chenepodiaceae 1 0.8 Ericaceae Rhododendron 3 2.3 Fabaceae Fabaceae 3 2.3 5 3.8 Fabaceae Astragalus 5 3.9 1 0.8 3 2.4 Fabaceae Hedysarum 6 4.7 3 2.3 Fabaceae Onobrychis 3 2.3 6 4.5 1 0.8 Fabaceae Trifolium 11 8.3 94 74.0 Fabaceae Vicia 4 3.0 Fagaceae Castanea sativa 1 0.8 1 0.8 Geraniaceae Geranium 1 0.8 Lamiaceae Lamiaceae 4 3.0 1 0.8 Lamiaceae Salvia 3 2.3 Onagraceae Epilobium 1 0.8 Papaveraceae Papaver 1 0.8 1 0.8 Plantaginaceae Plantago 3 2.3 6 4.7 Poaceae Poaceae 1 0.8 Polygonaceae Rumex 1 0.8 2 1,6 Ranunculaceae Ranunculaceae 2 1.5 Rosaceae Rosaceae 3 2.3 6 4.5 4 3.1 Rosaceae Potentilla 1 0.8 3 2.4 Total pollen count 129 132 127 The three honey samples contained 34 plant species from 20 different flowering plant families (Table 3). The most common families were Asteraceae, Apiaceae, Fabaceae, and Cistaceae. Asteraceae was the most common family in honey sample ZT1, while Fabaceae was the most common one in ZT2 and ZT3. Astragalus L. , Helianthemum Mill., and Onobrychis Mill. are the common taxa at the generic level in all honey samples, however, Campanula L. was the abundant taxa at the generic level in sample ZT1, while Trifolium Tourn. ex L. was abundant in samples ZT2 and ZT3. For each of the honey samples, Figure 4 shows the percentage of the top six taxa at the family level. Asteraceae was found to be the most predominant family in sample ZT1 accounting for 35.6% followed by Apiaceae (24.8%), Fabaceae (13.2%), Campanulaceae (9.3%), Brassicaceae (3.9%) and Cistaceae (3.1%). In sample ZT2, Fabaceae ranked first with 22.7%. The remaining families include Apiaceae (20.5%), Brassicaceae (16.7%), Asteraceae (12.2%), and Cistaceae (5.3%). The Fabaceae family accounted for 77.2% of sample ZT3, while families like Rosaceae (5.5%), Plantaginaceae (4.7%), Cistaceae (4,7%), Apiaceae (1.6%), and Asteraceae (1.6%) were found in minor quantities. Figure 4. The percentage (%) distribution of pollen detected according to melissopalynological data at family level. a) Sample ZT1, b) Sample ZT2, c) Sample ZT3 NGS findings Using the Illumina MiSeq, ITS1 and ITS2 tagged sequences were identified in a total of 350,580 sequences with a length of 251 bp. 74.4% of these sequences (260,885) were characterized to a family, genus, or species level. It was observed that ITS2 typically yielded fewer reads than ITS1. The number of ITS1 reads across all honey samples ranged from 52,375 to 154,485, whereas ITS2 reads ranged from 7,485 to 18,150. The average quality score for each honey sample is 37. Table 4 provides the percentages of Q20 and Q30 values based on Illumina data. The average QV20 and QV30 readings for each honey sample varied from 91% to 96% and 87% to 94%, respectively. Table 4. Summary of the Illumina results. ITS1 ITS2 ITS1 ITS2 ITS1 ITS2 Total reads 103885 14200 52375 18150 154485 7485 Reads (251 bp) 78690 10130 41055 11785 113910 5315 Unknown reads 25195 4070 11320 6365 40575 2170 Mean QV 37 37 37 37 37 37 Mean % QV20 reads 95.64 92 94 92 92 91 Mean % QV30 reads 93.45 88 92 89 88 87 Number identified to family, genus, or species level (%) 76 71 78 65 74 71 Number of taxa detected 28 29 28 35 39 21 A total of 310,745 filtered paired-end reads for ITS1 and 39,835 for ITS2 were obtained from sequencing data across three honey samples. Of the filtered reads, 233,655 (75.2%) for ITS1 and 27,230 (68.4%) for ITS2 matched the custom-prepared database at least at the family level. Additionally, all sequences were rechecked and supplemented through BLAST with specific filtering on NCBI. Reads corresponding to non-flowering plants and other organisms were filtered out and removed from the matching taxa. Table 5 shows the plant taxa identified across the three honey samples, revealing a total of 54 plant taxa from 27 plant families. Of these, 29 taxa were identified at the specific level, while others were classified only at the generic level. The families with the highest number of taxa present across all samples were Asteraceae, Fabaceae, Lamiaceae, and Rosaceae. The dominant taxa in the sample ZT1 at the generic and specific level are Artemisia L., Dianthus L., Salvia verticillata L., Cotoneaster Medik., Crataegus L., and Potentilla crantzii (Crantz) Fritsch. Astragalus fragrans Bunge, Salvia verticillata , Cotoneaster , Potentilla crantzii , and Prunus divaricata Ledeb. are dominant taxa in ZT2 honey sample. And the notable taxa in ZT3 include Taraxacum bessarabicum (Hornem.) Hand.-Mazz . , Betula litwinowii , Dianthus , Juglans regia L . , Potentilla crantzii , Prunus divaricata , Prunus domestica L . , and Verbascum L. (Table 5). Table 5. Identified plant taxa for the three honey samples using NGS: Dominant (red: >10000), Secondary (green: 1001-10000), Minor (dark yellow: 101-1000), and Trace (light yellow: 0-100). not-yet-known not-yet-known not-yet-known unknown ITS1 ITS2 ITS1 ITS2 ITS1 ITS2 Apiaceae Eryngium giganteum 29 16 Apiaceae Heracleum 34 Asteraceae Asteraceae 45 30 16 Asteraceae Achillea millefolium 91 Asteraceae Artemisia 1855 179 112 125 14 Asteraceae Centaurea 17 107 Asteraceae Helianthus tuberosus 29 89 30 Asteraceae Helichrysum pallasii 109 Asteraceae Lactuca racemosa 15 28 Asteraceae Taraxacum bessarabicum 29 146 46 1354 110 Berberidaceae Berberis vulgaris 154 15 47 96 Betulaceae Betula litwinowii 229 2942 Boraginaceae Echium vulgare 31 Brassicaceae Brassicaceae 114 893 154 30 1043 207 Caryophyllaceae Dianthus 9704 357 661 15 2615 Cistaceae Helianthemum nummularium 26 61 Convolvulaceae Calystegia sepium 357 168 93 83 Cucurbitaceae Cucurbita 89 15 78 Cyperaceae Carex 57 9 Ericaceae Rhododendron 30 Ericaceae Vaccinium 200 714 146 320 685 28 Fabaceae Astragalus fricki 314 89 31482 140 193 Fabaceae Hedysarum hedysaroides 57 45 15 31 Fabaceae Lotus corniculatus 156 Fabaceae Melilotus albus 45 Fabaceae Onobrychis 45 9 31 Fabaceae Phaseolus vulgaris 314 89 9 76 156 Fabaceae Trifolium 171 31 Fabaceae Vicia cracca 371 9 16 Fagaceae Quercus 76 529 Gentinanaceae Gentiana 536 62 Juglandaceae Juglans regia 143 179 26 427 50417 3883 Lamiaceae Mentha longifolia 343 45 30 498 Lamiaceae Salvia verticillata 4453 2187 1939 2653 1992 179 Lamiaceae Stachys 571 134 Lamiaceae Teucrium chamaedrys 29 89 34 30 31 55 Papaveraceae Papaver lateritium 60 Poaceae Poaceae 200 223 17 61 202 14 Polygonaceae Polygonum bistorta 200 Polygonaceae Rumex 91 16 28 Rosaceae Cotoneaster 1827 446 2934 107 872 55 Rosaceae Crataegus 6793 536 34 854 560 14 Rosaceae Potentilla 9647 120 168 3004 Rosaceae Potentilla crantzii 39616 1294 2273 3247 9371 14 Rosaceae Prunus divaricata 60 1829 24578 28 Rosaceae Prunus domestica 45 17 152 2926 28 Rosaceae Sorbus aucuparia 9 488 5650 Salicaceae Salix 9 15 62 28 Sapindaceae Acer 29 189 107 654 138 Scrophulariaceae Scrophularia 86 179 26 15 171 28 Scrophulariaceae Verbascum 1370 179 15 1977 Solanaceae Solanum tuberosum 114 45 Thymelaeaceae Daphne glomerata 29 893 343 61 560 69 Verbenaceae Verbena 89 61 62 Reads 78690 10130 41055 11785 113910 5315 Total number of reads (ITS1+ITS2) 88820 52840 119225 Figure 5 illustrates the distribution of the different families identified using NGS analysis. The top five families with the highest percentages are highlighted across all samples. The Rosaceae family represented 67.9 % of the sample ZT1. The remaining families, listed in descending order of percentages, are Caryophyllaceae (11.2%), Lamiaceae (8.9%), Asteraceae (2.5%), and Fabaceae (1.7%). In the ZT2 sample, Fabaceae occupied the foremost position with 59.8%. The remaining families include Rosaceae (23.3%), Lamiaceae (8.8%), Caryophyllaceae (1.3%), and Asteraceae (1.1%). In the sample ZT3, Juglandaceae was the most represented family (45.5%). The remaining families are Rosaceae (39.5%), Lamiaceae (2.3%), Caryophyllaceae (2.2%), and Asteraceae (1.7%). Figure 5. Percentage (%) distribution of pollen detected according to NGS data at family level. a) Sample ZT1, b) Sample ZT2, c) Sample ZT3 A comparison of the common families between ITS1 and ITS2 across all samples is demonstrated in Figure 6. Of the 20 families identified in sample ZT1, 12 families (60%) were shared between ITS1 and ITS2. Correspondingly, ZT2 and ZT3 exhibited 63% and 61% shared families, respectively, between ITS1 and ITS2 (Figure 6). Figure 6. Vann diagram of the shared families between ITS1 and ITS2 for the three honey samples. Comparison of melissopalynological and NGS data NGS analysis identified 37% more plant taxa in all honey samples compared to the melissopalynological method. While several families were identified using both methods (43.75%), others were identified using either NGS (37.50%) or melissopalynology (18.75%). According to NGS analysis, 37 different taxa (8.1% at the family, 43.2% at generic, and 48.6% at specific level) were identified for ZT1, however, 15 different taxa (40% at the family, 53.3% at generic and 6.7 % at specific level) were identified using melissopalynology (Table 6). Additionally, when comparing the results of the two methods, 48 taxa (Family: 14.6%, genus: 47.9%, and species: 37.5%) were found in NGS or melissopalynology. Similarly, 4 taxa (Family: 50%, genus: 50%, and species: 0%) were found common in both methods (Table 6). Table 6. Comparison of NGS and melissopalynology results. NGS Taxa (n) 37 45 42 Family (%) 8.1 6.7 7.1 Genus (%) 43.2 42.2 40.5 Species (%) 48.6 51.1 52.4 Melissopalynology Taxa (n) 15 29 14 Family (%) 40.0 27.6 42.9 Genus (%) 53.3 69.0 57.1 Species (%) 6.7 3.4 0.0 NGS or Melissopalynology Taxon (n) 48 66 50 Family (%) 14.6 13.6 14.0 Genus (%) 47.9 50.0 42.0 Species (%) 37.5 36.4 44.0 NGS and Melissopalynology Taxon (n) 4 8 6 Family (%) 50.0 25.0 33.3 Genus (%) 50.0 75.0 66.7 Species (%) 0.0 0.0 0.0 29 plant families including more than one genus or species are recorded for NGS or melissopalynology (Table 7). Of these, three families have the same number of taxa detected for DNA and microscopy. However, more taxa were identified by DNA metabarcoding for Rosaceae, Asteraceae, Fabaceae, and Lamiaceae. For Boraginaceae, Campanulaceae, Cistaceae, and Ranunculaceae, microscopy detected more taxa represented by only one or two pollen grains. The proportion of taxa that could be distinguished at both specific and generic levels was higher in DNA metabarcoding. In all families, DNA metabarcoding detected 54 taxa (at the family, generic, and specific), while in melissopalynology 27 taxa (at the family, generic, and specific) were detected. Of the 54 taxa identified using DNA metabarcoding, 23 were distinguished at the genus and 26 at the species level, whereas using microscopy, 23 of the 27 taxa were distinguished at the genus and only 1 at the species level (Figure 7). Figure 7 . The number of taxa (generic and specific) detected within each family in the three honey samples analyzed by Illumina (I), melissopalynology (M), a combination of both techniques (I or M), and both techniques in common (I & M). A total of 14 families and 17 genera were found common to both techniques. These common taxa are shown in Figure 8. Figure 8. Distribution of common taxa (family, generic, specific) identified as a result of microscopic and NGS analysis in the analyzed honey samples. Although the results of the two methods showed variation, both approaches contained the dominant taxa in varying relative abundances providing 23% agreement between NGS and melissopalynology (Figure 8). Taxa showing agreement at the family level are Asteraceae, Brassicaceae, and Poaceae. Taxa showing similarity at the genus level are Artemisia, Centeaurea, Dianthus, Onobrychis, Potentilla , and Rumex . No similarity was found at the species level (Table 7). In addition, a total of 6 families (Apiaceae, Asteraceae, Caryophyllaceae, Fabaceae, Lamiaceae, and Rosaceae) were found common to both methods. Table 7. Plant taxa identified by melissopalynology (M) and Illumina (I) analysis in the three honey samples (percentages of total pollen grains and DNA sequences). The red outline indicates the common taxa between the two methods. M I M I M I Apiaceae Apiaceae 24.81 20.45 1.57 Apiaceae Eryngium giganteum 0.03 0.06 0.01 Apiaceae Heracleum 0.06 Asteraceae Asteraceae 33.33 0.05 5.30 0.06 1.57 0.01 Asteraceae Achillea millefolium 0.17 Asteraceae Artemisia 2.30 2.29 0.21 0.12 Asteraceae Centaurea 3.80 0.23 Asteraceae Helianthus 0.80 Asteraceae Helianthus tuberosus 0.13 0.06 Asteraceae Helichrysum pallasii 0.09 Asteraceae Lactuca racemosa 0.03 0.02 Asteraceae Taraxacum 2.30 Asteraceae Taraxacum bessarabicum 0.03 0.36 1.23 Berberidaceae Berberis vulgaris 0.32 0.12 Betulaceae Betula litwinowii 0.43 2.47 Boraginaceae Echium 1.50 Boraginaceae Echium vulgare 0.03 Boraginaceae Myosotis 1.60 Brassicaceae Brassicaceae 3.88 1.13 1.,67 0.35 1.05 Campanulaceae Campanula 9.30 1.52 Caryophyllaceae Caryophyllaceae 3.10 0.79 Caryophyllaceae Dianthus 11.20 1.50 1.8 2.19 Chenepodiaceae Chenepodiaceae 0.76 Cistaceae Cistus 2.30 Cistaceae Helianthemum 3.10 2.30 4.70 Cistaceae Helianthemum nummularium 0.16 Convolvulaceae Calystegia sepium 0.40 0.32 0.15 Cucurbitaceae Cucurbita 0.10 0.03 0.07 Cyperaceae Carex 0.06 0.02 Ericaceae Rhododendron 2.30 0.06 Ericaceae Vaccinium 1.03 0.88 0.60 Fabaceae Fabaceae 2.33 3.79 Fabaceae Astragalus 3.90 0.80 2.40 Fabaceae Astragalus fricki 0.45 59.58 0.28 Fabaceae Hedysarum 4.70 2.0 Fabaceae Hedysarum hedysaroides 0.11 0.03 0.03 Fabaceae Lotus corniculatus 0.13 Fabaceae Melilotus albus 0.05 Fabaceae Onobrychis 2.30 0.05 4.50 0.02 0.80 0.03 Fabaceae Phaseolus vulgaris 0.45 0.16 0.13 Fabaceae Trifolium 0.19 8.30 74.00 0.03 Fabaceae Vicia 3.00 Fabaceae Vicia cracca 0.42 0.02 0.01 Fagaceae Castanea sativa 0.80 0.80 Fagaceae Quercus 0.14 0.44 Gentinanaceae Gentiana 0.60 0.05 Geraniaceae Geranium 0.80 Juglandaceae Juglans regia 0.36 0.86 45.54 Lamiaceae Lamiaceae 3.03 0.79 Lamiaceae Mentha longifolia 0.44 0.06 0.42 Lamiaceae Salvia 2.30 Lamiaceae Salvia verticillata 7.49 8.69 1.82 Lamiaceae Stachys 0.79 Lamiaceae Teucrium chamaedrys 0.13 0.12 0.07 Onagraceae Epilobium 0.80 Papaveraceae Papaver 0.80 0.80 Papaveraceae Papaver lateritium 0.11 Plantaginaceae Plantago 2.30 4.70 Poaceae Poaceae 0.48 0.15 0.79 0.18 Polygonaceae Polygonum bistorta 0.23 Polygonaceae Rumex 0.80 0.17 1.60 0.04 Ranunculaceae Ranunculaceae 1.52 Rosaceae Rosaceae 2.33 4.55 3.15 Rosaceae Cotoneaster 2.56 5.75 0.78 Rosaceae Crataegus 8.26 1.68 0.48 Rosaceae Potentilla 10.86 0.80 0.55 2.40 2.52 Rosaceae Potentilla crantzii 46.06 10.45 7.87 Rosaceae Prunus divaricata 3.57 20.64 Rosaceae Prunus domestica 0.05 0.32 2.48 Rosaceae Sorbus aucuparia 0.94 4.74 Salicaceae Salix 0.05 0.08 Sapindaceae Acer 0.03 0.56 0.66 Scrophulariaceae Scrophularia 0.30 0.08 0.17 Scrophulariaceae Verbascum 1.75 0.03 1.66 Solanaceae Solanum tuberosum 0.18 Thymelaeaceae Daphne glomerata 1.04 0.76 0.53 Verbenaceae Verbena 0.10 0.12 0.05 NGS analysis identified Asteraceae, Fabaceae, and Rosaceae as dominant and secondary families, while microscopic analysis identified Asteraceae, Apiaceae, and Fabaceae as dominant and secondary families (Table 7). Furthermore, some families (Berberidaceae, Betulaceae, Convolvulaceae, Cucurbitaceae, Cyperaceae, Gentinanaceae, Juglandaceae, Salicaeae, Sapindaceae, Scrophulariaceae, Solanaceae, Thymelaeaceae, Verbenaceae), which were not found by the melissopalynological data, were identified by the NGS analysis. A comparison of the two methods showed 29%, 33%, and 26% common families for the three samples, respectively (Figure 9). not-yet-known not-yet-known not-yet-known unknown Figure 9. Number of families identified by NGS and melissopalynology for the three honey samples. The Spearman’s correlation coefficients for the relationships between melissopalynology and ITS, between melissopalynology and ITS1, between melissopalynology and ITS2, and between ITS1 and ITS across each sample are shown in Table 8. Overall, no significant correlation was found between melissopalynology and NGS for any of the samples. At the family level, a weak positive correlation was found between the microscopy and ITS1 for samples ZT1 and ZT2. However, none of the samples showed any correlation between microscopy and ITS2. At the taxa level, a weak negative correlation was found between the two methods for samples ZT1 and ZT2. A moderate positive correlation was found between the ITS1 and ITS2 regions at both the family and the taxa levels across all samples (Table 8). Table 8. Spearman’s correlation coefficient for the honey samples (M: Mellisopalynolgy, ITS: Internal Transcribed Spacer) Spearman’s (ρ) P-value Spearman’s (ρ) P-value ZT1 M & ITS 0.290 0.100 -0.207 0.071 M & ITS1 0.359 0.040 -0.234 0.040 M & ITS2 0.196 0.270 -0.121 0.294 ITS1 & ITS2 0.630 0.000 0.620 0.000 ZT2 M & ITS 0.247 0.166 -0.416 0.000 M & ITS1 0.360 0.040 -0.303 0.007 M & ITS2 0.059 0.750 -0.352 0.002 ITS1 & ITS2 0.483 0.000 0.520 0.000 ZT3 M & ITS 0.080 0.657 -0.175 0.129 M & ITS1 0.064 0.720 -0.167 0.148 M & ITS2 0.166 0.360 -0.174 0.125 ITS1 & ITS2 0.664 0.000 0.580 0.000 not-yet-known not-yet-known not-yet-known unknown The two techniques were also compared using the Diversity (H’) and Evenness indices (J’) at the family level as well as across all identified taxonomic levels. The value of the maximum diversity (Hmax) for the honey samples across all identified taxonomic levels ranged between 2.64 and 3.81 (Figure 10). Sample ZT2M exhibited the highest diversity (2.85 for Hmax = 3.36) followed by ZT1M (2.09 for Hmax = 2.71), and ZT1I (2.00 for Hmax = 3.61). ZT3M had the lowest diversity of 1.19 for Hmax = 2.64. Similarly, the maximum and the minimum values of the evenness indices for the honey samples at the taxa level ranged between 0.44 and 0.85 for ZT3M and ZT1M, respectively. For the NGS results, the evenness indices for the samples ZT1I, ZT2I, and ZT3I were 0.55, 0.44, and 0.53, respectively. Figure 10. Shannon-Wiener and Pielou Evenness Indices for the two techniques at the taxon level (a, b) and the family level (c, d). The letters M and I at the suffix of each sample name indicates Melissopalynology and Illumina respectively. At the family level, ZT2M revealed the maximum diversity (2.28 for H max = 2.94), while ZT3M had the lowest diversity (1.0 for H max = 2.4). ZT2M and ZT2I had the highest (0.77) and the lowest (0.40) evenness index at the family level. The R-coefficient of the three honey samples at the family level is given in Figure 11. Red bars indicate overrepresented families, while the blue bars indicate underrepresented families. In the sample ZT1, Rosaceae (29.15) is the most overrepresented family, i.e., Rosaceae is 29 times more represented in NGS results than in melissopalynology. Similarly, Apiaceae is the least represented (0.0013) family, i.e., Apiaceae is 769 times more represented in melissopalynology than NGS. Caryophyllaceae (0.85) from sample ZT2 and Asteraceae (0.93) from sample ZT3 were the only two families that were almost equally represented in both methods (Figure 11). not-yet-known not-yet-known not-yet-known unknown Figure 11. R-coefficient showing the over and underrepresented taxa. not-yet-known not-yet-known not-yet-known unknown 1 Discussion The current findings show that NGS is an alternative method to determine the floral sources and geographical origin of Anzer honey. It can also provide insights into the floral resources available to honeybees in the surrounding environment. However, the lack of complete floral reference databases and lower taxonomic resolution for many of the taxa makes this method contingent on other identification methods. Nevertheless, NGS can be considered a more thorough and dependable methodology when compared to traditional melissopalynology. This section involves a detailed discussion of the melissopalynology and NGS results of Anzer honey based on several statistical and ecological indices. Louveaux et al. (1978) characterized honey as monofloral if over 45% of its pollen originates from a single species. Samples ZT1 and ZT2 exhibited diversity in the pollen grains for each species not exceeding the 45% threshold rendering them to be classified as multifloral honey by Malkoç et al. (2019). However, 74% of ZT3’s pollen grains were from Trifolium, indicating that it is monofloral despite its initial classification. This suggests that the honey was produced in a region or season with abundant clover or that the bees were feeding primarily on clover pollen. Similar findings have been reported in other studies. For instance, Sorkun & Doğan (1995) reported that Trifolium pollen grains are dominant in honey samples from the Anzer region. In line with Sorkun & Doğan (1995) results, Güner et al. (1987) reported Trifolium pollen as dominant in honey from Rize. Sorkun & Doğan (1995) stated that the main distinguishing feature of the Anzer Honey is the presence of Myosotis pollen in minor (3%-15%) or rare (less than 3%) quantities. In the current study, the Myosotis was identified in only sample ZT1 by microscopy. The present authors contend that Myosotis in itself should not be considered the sole distinguishing feature for identifying Anzer honey. Basing the identification of Anzer honey solely on the presence of Myosotis pollen grains may lead to mislabeling due to natural heterogeneity of Anzer honey. At the same time, Myosotis is not recorded by Illumina methods in any of the examined honey samples contrary to Sorkun & Doğan (1995). According to Hawkins et al. (2015), the identification of the members belonging to Boraginaceae family using metabarcoding is very difficult. So, this limitation in metabarcoding accuracy further complicates the use of Myosotis as a definitive marker for Anzer honey. The sequencing results revealed a total of 350,580 ITS1 and ITS2 reads (averaging 251 bp in length) from the honey metabarcoding analysis, with 74.4% of these reads successfully classified at the family, genus, or species level. These findings are consistent with findings documented in the literature (Keller et al., 2015; Kraaijeveld et al., 2015). Moreover, although ITS1 yielded a higher overall read count, ITS2 demonstrated greater species diversity despite having lower read counts, particularly in the ZT1 and ZT2 samples. As a result, the ITS2 marker emerged as the preferred option for achieving higher taxonomic resolution which is in accordance with the literature findings (Chen et al., 2010). The discrepancies between the reference database and the metabarcoding results may be due to the use of only one genetic marker (ITS in this case) instead of integrating several genetic markers (e.g., rbc L and mat K), which improves taxonomic precision but requires a comprehensive reference library. For example, although Swertia iberica Fisch. & C.H. May was sampled in the Anzer region, it could not be identified in any honey samples due to the absence of a reference sequence for this taxon in the database; instead, the metabarcoding sequence matched to the genus Gentiana within the same family. The current NGS results revealed a high level of agreement with the microscopic findings for many of the most dominant taxa. Both methods identified 14 common families, representing 70% of those detected by melissopalynology, and 17 common genera, accounting for 74% of the genera observed under microscopy. The discrepancies between the two methods can be explained by the differing resolutions each technique provides. Additionally, some studies suggest that these variations may also stem from differences in sampling procedures and the heterogeneous nature of honey. For instance, while 10 g of the sample was used for melissopalynology, DNA isolation required 50 g of honey (Hawkins et al., 2015). When the data from both techniques are compared, it is hard to distinguish similarities or differences at the specific level. In a similar study, Hawkins et al., (2015) identified many taxa at the family level with microscopy, while the same taxa were identified at the specific level with NGS, which led to different results. They observed that metabarcoding has a greater degree of discrimination power, which makes it challenging to compare the results directly to those obtained through microscopy. For example, in the melissopalynological data of the current study, Apiaceae could only be identified at the family level, whereas it was identified at the generic level (Heracleum ) and the specific (Eryngium giganteum ) levels using metabarcoding. Another family, Rosaceae, was only identified at the generic level (Potentilla) in the melissopalynological data, while Cotoneaster, Crataegus, Potentilla crantzii, Prunus divaricata, Prunus domestica and Sorbus aucuparia taxa were identified in the NGS data. Another reason for the variation in results could be attributed to the quantity of taxa present in the examined samples. Certain species, especially those in low quantities, can be missed when a mixed sample is amplified (Gibson et al., 2014; Hajibabaei et al., 2011). This indicates that species identified from a single pollen grain using microscopy in the current study (Epilobium, Geranium, and Castanea sativa ) are less probable to be discovered via DNA metabarcoding. Moreover, potential errors during amplification and sequencing further restrict the precise identification of the samples. Özkök et al., (2023) compared melissopalynology and metabarcoding by Ion Torrent using rbc L and trn H-psb A plastid barcode markers on 40 monofloral honey samples from different geographical regions of Türkiye. Three of these honey samples belonged to Rize-Anzer honey. The findings analyzed by Özkök et al. (2023) revealed that 18 families were common to both techniques. Among these, 10 families (Apiaceae, Asteraceae, Brassicaceae, Caryophyllaceae, Ericaceae, Fabaceae, Fagaceae, Lamiaceae, Poaceae, and Rosaceae) were also identified in the current study. Although the current study was based on only a few honey samples, the results support the potential to detect fraud or mislabeling of honey by comparing local plant taxa with those detected in the honey sample. Spearman’s correlation analysis of the ITS regions (ITS1/ITS2) and the melissopalynological data revealed a moderate linear relationship for samples ZT1 and ZT2, with a positive association at the family level and a negative association when assessed across all identified taxonomic ranks. However, no statistically significant correlation was observed for sample ZT3, either when evaluated at the family level or across all identified taxonomic ranks. Similarly, Richardson et al. (2015a) reported no significant relationship between the two methods based on the ITS2 barcode region. However, in consecutive studies in the same year, when combining different barcode regions (ITS2, mat K, rbc L), Richardson et al. (2015b) reported a statistically significant relationship between the two methods. While metabarcoding identified a greater diversity of taxa, its Shannon and Pielou indices were unexpectedly lower compared to melissopalynology. This disparity likely stems from the disproportionately high read counts assigned to a few taxa (Astragalus fricki, Juglans regia, Potentilla crantzii, and Prunus divaricata ) in the metabarcoding data, which skews the indices and highlights the quantitative unreliability (Bell et al., 2019; Milla et al., 2021) of NGS read counts. This unreliability is further highlighted by the R-coefficient values of the two methods. With the exception of Caryophyllaceae in ZT2 (R-value = 0.85) and Asteraceae in ZT3 (R-value = 0.93), all the families in all the samples were either underrepresented or overrepresented (Figure 11). Despite these limitations, the indices provide valuable qualitative insights. For microscopy data, ZT2 showed the highest diversity (H’ = 2.85, J’ = 0.85), indicating a balanced distribution of taxa, while ZT3 had the lowest diversity (H’ = 1.19, J’ = 0.45) due to Trifolium dominance. For NGS data, ZT2 exhibited the lowest diversity (H’ = 2.00, J’ = 0.44), primarily driven by Astragalus frickii dominance. These findings reinforce the caution needed when interpreting quantitative results from metabarcoding while still underscoring its utility for detecting broader taxonomic diversity. 1 Discussion not-yet-known not-yet-known not-yet-known unknown 1 Conclusions This study involved a comparative analysis of three Anzer honey samples based on the findings of traditional melissopalynology and metabarcoding. Recently both methods have been used to determine the botanical and geographical origin of the honey samples. Our results revealed that metabarcoding surpassed melissopalynology in identifying more species and can potentially serve as a standalone method for honey botanical profiles. However, both methods look at different dimensions to determine the diversity of pollen grains in honey and have potential weaknesses. Therefore, researchers should keep these aspects in mind, and it is better to focus on both methods to maximize the identification of pollen grains at family and/or specific levels in honey. The main conclusions reached in this study are presented below in headings; The microscopic analysis of the three honey samples revealed 34 plant taxa belonging to 20 different families. Of the 20 families, six were only identified by melissopalynology. Most of the taxa were identified at the generic level. A total of 54 plant taxa belonging to 27 families (13 of which were only identified by metabarcoding) were recorded using ITS1/ITS2 region metabarcoding. 29 of the 54 taxa were identified at the species level, demonstrating metabarcoding as a superior identification method to melissopalynology. Asteraceae, Fabaceae, Lamiaceae, and Rosaceae were the families with the highest number of taxa detected in all the samples. The NGS analysis detected 37% more plant taxa than the melissopalynological method for all honey samples. Due to the higher resolution provided by the metabarcoding, the commonality between the two methods could only be actualized at the family and the genus level. Although there were some differences in the plant taxa identified by both techniques, the dominant floral components identified showed high similarity. Spearman’s statistical correlation revealed a moderate linear relation between the two methodologies, which in the future can be improved by incorporating other markers like rbc L and trn H-psb A. Ecological diversity analyses like Shannon-Wiener and Pielou’s Evenness indices results for metabarcoding were unexpectedly lower compared to melissopalynology highlighting the quantitative unreliability of metabarcoding. 1 Conclusions AUTHOR CONTRIBUTIONS Zeynep TÜRKER: data curation (lead), formal analysis (lead), methodology (equal), sampling (equal), software (lead), validation (equal), writing – original draft (lead), writing – review and editing (equal). Kamil COŞKUNÇELEBİ: conceptualization (lead), methodology (equal), project administration (lead), resources (lead), sampling (equal), supervision (lead), validation (equal), writing – original draft (equal), writing – review and editing (equal). Murat Erdem GÜZEL: methodology (equal), sampling (equal), software (equal), validation (equal), writing – review and editing (equal). Serdar MAKBUL: methodology (equal), resources (equal), sampling (equal), validation (equal), writing – review and editing (equal). ACKNOWLEDGEMENTS This study was partially supported by Karadeniz Technical University. The authors thank Prof. Dr. Mahir BUDAK from Sivas Cumhuriyet University’s Advanced Technology Research and Application Center (CÜTAM) for providing the necessary help in conducting the metabarcoding analysis, and Dr. Esra DEMIR KANBUR from Central Research Laboratory Recep Tayyip Erdoğan University for their expertise in melissopalynological analysis. The authors would also like to express their gratitude to the S.S. Anzer Ballıköy Agricultural Development Cooperative for providing the honey samples. CONFLICT OF INTEREST STATEMENT The authors declare no conflict of interest. DATA AVAİLABİLİTY STATEMENT The data that supports the findings of the study are available from the corresponding author upon reasonable request. References 1. Anonymous. (2014). Ulusal Biyolojik Çeşitlilik Envanter ve İzleme Proje Raporları . Balkanska, R., Stefanova, K., Stoikova-Grigorova, R., & Ignatova, M. (2020). A preliminary assessment of trnH-psbA as DNA barcode for botanical identification of polyfloral honey samples and comparison with rbcL marker. Bulgarian Journal of Agricultural Science , 26 (1), 238–242. Bell, K. L., Burgess, K. S., Botsch, J. C., Dobbs, E. K., Read, T. D., & Brosi, B. J. (2019). Quantitative and qualitative assessment of pollen DNA metabarcoding using constructed species mixtures. Molecular Ecology , 28 (2), 431–455. https://doi.org/10.1111/MEC.14840 Bell, K. L., De Vere, N., Keller, A., Richardson, R. T., Gous, A., Burgess, K. S., & Brosi, B. J. (2016). Pollen DNA barcoding: Current applications and future prospects. Genome , 59 (9), 629–640. https://doi.org/10.1139/GEN-2015-0200/ASSET/IMAGES/GEN-2015-0200TAB1.GIF Benson, D. A., Clark, K., Karsch-Mizrachi, I., Lipman, D. J., Ostell, J., & Sayers, E. W. (2015). GenBank. Nucleic Acids Research , 43 (Database issue), D30. https://doi.org/10.1093/NAR/GKU1216 Bruni, I., Galimberti, A., Caridi, L., Scaccabarozzi, D., De Mattia, F., Casiraghi, M., & Labra, M. (2015). A DNA barcoding approach to identify plant species in multiflower honey. Food Chemistry , 170 , 308–315. https://doi.org/10.1016/J.FOODCHEM.2014.08.060 Chen, S., Yao, H., Han, J., Liu, C., Song, J., Shi, L., Zhu, Y., Ma, X., Gao, T., Pang, X., Luo, K., Li, Y., Li, X., Jia, X., Lin, Y., & Leon, C. (2010). Validation of the ITS2 Region as a Novel DNA Barcode for Identifying Medicinal Plant Species. PLOS ONE , 5 (1), e8613. https://doi.org/10.1371/JOURNAL.PONE.0008613 Cheng, H., Jin, W., Wu, H., Wang, F., You, C., Peng, Y., & Jia, S. (2007). Isolation and PCR Detection of Foreign DNA Sequences in Bee Honey Raised on Genetically Modified Bt (Cry 1 Ac) Cotton. Food and Bioproducts Processing , 85 (2), 141–145. https://doi.org/10.1205/FBP06056 Çil, E. (2023). A Functional Food: Anzer Honey. In Versatile Approaches to Engineering and Applied Sciences: Materials and Methods . Özgür Yayınları. Eteraf-Oskouei, T., & Najafi, M. (2013). Traditional and Modern Uses of Natural Honey in Human Diseases: A Review. Iranian Journal of Basic Medical Sciences , 16 (6), 731–742. www.mums.ac.ir/basic_medical/en/index Galimberti, A., De Mattia, F., Bruni, I., Scaccabarozzi, D., Sandionigi, A., Barbuto, M., Casiraghi, M., & Labra, M. (2014). A DNA Barcoding Approach to Characterize Pollen Collected by Honeybees. PLOS ONE , 9 (10), e109363. https://doi.org/10.1371/JOURNAL.PONE.0109363 Gibson, J., Shokralla, S., Porter, T. M., King, I., Van Konynenburg, S., Janzen, D. H., Hallwachs, W., & Hajibabaei, M. (2014). Simultaneous assessment of the macrobiome and microbiome in a bulk sample of tropical arthropods through DNA metasystematics. Proceedings of the National Academy of Sciences of the United States of America , 111 (22), 8007–8012. https://doi.org/10.1073/PNAS.1406468111/SUPPL_FILE/PNAS.1406468111.SAPP.PDF Güner, A., Aslan, S., Ekim, T., Vural, M., & Babaç, M. T. (2012). Türkiye Bitkileri Listesi (Damarlı Bitkiler) . Nezahat Gökyiğit Botanik Bahçesi ve Flora Araştırmaları Derneği Yayını. Güner, Vural, M., & Sorkun, K. (1987). Rize florası, vejetasyonu ve yöre ballarının polen analizi . Hajibabaei, M., Shokralla, S., Zhou, X., Singer, G. A. C., & Baird, D. J. (2011). Environmental Barcoding: A Next-Generation Sequencing Approach for Biomonitoring Applications Using River Benthos. PLOS ONE , 6 (4), e17497. https://doi.org/10.1371/JOURNAL.PONE.0017497 Hawkins, J., de Vere, N., Griffith, A., Ford, C. R., Allainguillaume, J., Hegarty, M. J., Baillie, L., & Adams-Groom, B. (2015). Using DNA Metabarcoding to Identify the Floral Composition of Honey: A New Tool for Investigating Honey Bee Foraging Preferences . https://doi.org/10.1371/journal.pone.0134735 Hepsağ, F. (2019). Determination of Total Phenolic Compounds and Antioxidant Capacity of Anzer Honey Produced in Rize, Turkey. Gıda The Journal of Food , 44 (4), 641–653. https://doi.org/10.15237/gida.GD19046 Hotaman, H. (2015). Anzer bal ve poleninin bazı biyoaktif özelliklerinin in vitro olarak incelenmesi . Recep Tayyip Erdoğan Üniversitesi. Keller, A., Danner, N., Grimmer, G., Ankenbrand, M., von der Ohe, K., von der Ohe, W., Rost, S., Härtel, S., & Steffan-Dewenter, I. (2015). Evaluating multiplexed next-generation sequencing as a method in palynology for mixed pollen samples. Plant Biology , 17 (2), 558–566. https://doi.org/10.1111/PLB.12251 Kraaijeveld, K., de Weger, L. A., Ventayol García, M., Buermans, H., Frank, J., Hiemstra, P. S., & den Dunnen, J. T. (2015). Efficient and sensitive identification and quantification of airborne pollen using next-generation DNA sequencing. Molecular Ecology Resources , 15 (1), 8–16. https://doi.org/10.1111/1755-0998.12288 Laube, I., Hird, H., Brodmann, P., Ullmann, S., Schöne-Michling, M., Chisholm, J., & Broll, H. (2010). Development of primer and probe sets for the detection of plant species in honey. Food Chemistry , 118 (4), 979–986. https://doi.org/10.1016/J.FOODCHEM.2008.09.063 Louveaux, J., Maurizio, A., & Vorwohl, G. (1978). Methods of Melissopalynology. Bee World , 59 (4), 139–157. https://doi.org/10.1080/0005772X.1978.11097714 Malkoç, M., Çakir, H., Kara, Y., Can, Z., & Kolaylı, S. (2019). Phenolic Composition and Antioxidant Properties of Anzer Honey from Black Sea Region of Turkey. DergiPark , 19 , 143–151. https://doi.org/10.31467/uluaricilik.602906 Milla, L., Sniderman, K., Lines, R., Mousavi-Derazmahalleh, M., & Encinas-Viso, F. (2021). Pollen DNA metabarcoding identifies regional provenance and high plant diversity in Australian honey. Ecology and Evolution , 11 (13), 8683–8698. https://doi.org/10.1002/ECE3.7679 Olivieri, C., Marota, I., Rollo, F., & Luciani, S. (2012). Tracking Plant, Fungal, and Bacterial DNA in Honey Specimens*. Journal of Forensic Sciences , 57 (1), 222–227. https://doi.org/10.1111/J.1556-4029.2011.01964.X Özkök, A., Akel Bilgiç, H., Kosukcu, C., Arık, G., Canlı, D., Yet, İ., & Karaaslan, C. (2023). Comparing the melissopalynological and next generation sequencing (NGS) methods for the determining of botanical origin of honey. Food Control , 148 , 109630. https://doi.org/10.1016/J.FOODCONT.2023.109630 PalDat. (2023). Palynological Database . https://www.paldat.org/ Pollen-Wiki. (2025). Pollen-Wiki . Pollen-Wiki. https://pollen.tstebler.ch/MediaWiki/index.php?title=Pollenatlas#gsc.tab=0 Richardson, R. T., Lin, C.-H., Quijia, J. O., Riusech, N. S., Goodell, K., & Johnson, R. M. (2015a). Rank-based characterization of pollen assemblages collected by honey bees using a multi-locus metabarcoding approach. Applications in Plant Sciences , 3 (11), 1500043. https://doi.org/10.3732/APPS.1500043 Richardson, R. T., Lin, C.-H., Sponsler, D. B., Quijia, J. O., Goodell, K., & Johnson, R. M. (2015b). Application of ITS2 metabarcoding to determine the provenance of pollen collected by honey bees in an agroecosystem. Applications in Plant Sciences , 3 (1), 1400066. https://doi.org/10.3732/APPS.1400066 Rize İl Tarım ve Orman Müdürlüğü. (2019). Anzer Balı (Patent 676). Türk Patent. Schnell, I. B., Fraser, M., Willerslev, E., & Gilbert, M. T. P. (2010). Characterization of insect and plant origins using DNA extracted from small volumes of bee honey. Arthropod-Plant Interactions , 4 (2), 107–116. https://doi.org/10.1007/S11829-010-9089-0 Smart, M. D., Cornman, R. S., Iwanowicz, D. D., McDermott-Kubeczko, M., Pettis, J. S., Spivak, M. S., & Otto, C. R. V. (2017). A Comparison of Honey Bee-Collected Pollen From Working Agricultural Lands Using Light Microscopy and ITS Metabarcoding. Environmental Entomology , 46 (1), 38–49. https://doi.org/10.1093/EE/NVW159 Sorkun. (1985). Balda Polen Analizi. Teknik Arıcılık Dergisi , 1 , 28–30. Sorkun, & Doğan, C. (1995). Pollen Analysis of Rize-Anzer (Turkish) Honey. Apiacta , 3 . Sorkun, K. (2008). Türkiye’nin Nektarlı Bitkileri, Polenleri ve Balları. In Palme Yayıncılık . Palme Yayıncılık. Terzioğlu, S. (1994). Of-İkizdere-Anzer Vadisi Florası . Karadeniz Teknik Üniversitesi. Ulusoy, E., & Kolaylı, S. (2014). Phenolic Composition and Antioxidant Properties of Anzer Bee Pollen. Journal of Food Biochemistry , 38 (1), 73–82. https://doi.org/10.1111/JFBC.12027 Ulusoy, E., Kolaylı, S., & Sarikaya, A. O. (2010). Antioxidant and Antimicrobial Activity of Different Floral Origin Honeys from Turkiye. Journal of Food Biochemistry , 34 (SUPPL. 1), 321–335. https://doi.org/10.1111/J.1745-4514.2009.00332.X Utzeri, V. J., Ribani, A., Schiavo, G., Bertolini, F., Bovo, S., & Fontanesi, L. (2018). Application of next generation semiconductor based sequencing to detect the botanical composition of monofloral, polyfloral and honeydew honey. Food Control , 86 , 342–349. https://doi.org/10.1016/J.FOODCONT.2017.11.033 Vorwohl, G. (1967). The microscopic analysis of honey, a comparison of its methods with those of the other branches of palynology. Review of Palaeobotany and Palynology , 3 (1–4), 287–290. https://doi.org/10.1016/0034-6667(67)90061-9 Wilson, E. E., Sidhu, C. S., Levan, K. E., & Holway, D. A. (2010). Pollen foraging behaviour of solitary Hawaiian bees revealed through molecular pollen analysis. Molecular Ecology , 19 (21), 4823–4829. https://doi.org/10.1111/J.1365-294X.2010.04849.X Wodehouse, R. P. (1935). Pollen grains and worlds of different sizes. The Scientific Monthly , 40 (1), 58–62. Crossref Google Scholar Information & Authors Information Version history V1 Version 1 28 February 2025 Copyright This work is licensed under a Non Exclusive No Reuse License. Keywords anzer honey illumina melissopalynology Authors Affiliations Zeynep Türker 0009-0004-8666-7793 [email protected] Karadeniz Technical University View all articles by this author Kamil Coskuncelebi 0000-0001-6432-9807 Karadeniz Technical University View all articles by this author Murat Güzel Karadeniz Technical University View all articles by this author Serdar Makbul Recep Tayyip Erdogan Universitesi View all articles by this author Metrics & Citations Metrics Article Usage 355 views 155 downloads .FvxKWukQNSOunydq8rnd { width: 100px; } Citations Download citation Zeynep Türker, Kamil Coskuncelebi, Murat Güzel, et al. Comparison of Next-Generation Sequencing and Traditional Melissopalynological Methods for Geographically Labeled Anzer Honey. Authorea . 28 February 2025. DOI: https://doi.org/10.22541/au.174074244.46911923/v1 If you have the appropriate software installed, you can download article citation data to the citation manager of your choice. Simply select your manager software from the list below and click Download. For more information or tips please see 'Downloading to a citation manager' in the Help menu . Format Please select one from the list RIS (ProCite, Reference Manager) EndNote BibTex Medlars RefWorks Direct import Tips for downloading citations document.getElementById('citMgrHelpLink').addEventListener('click', function() { popupHelp(this.href); return false; }); $(".js__slcInclude").on("change", function(e){ if ($(this).val() == 'refworks') $('#direct').prop("checked", false); $('#direct').prop("disabled", ($(this).val() == 'refworks')); }); View Options View options PDF View PDF Figures Tables Media Share Share Share article link Copy Link Copied! Copying failed. Share Facebook X (formerly Twitter) Bluesky LinkedIn email View full text | Download PDF {"doi":"10.22541/au.174074244.46911923/v1","type":"Article"} Now Reading: Share Figures Tables Close figure viewer Back to article Figure title goes here Change zoom level Go to figure location within the article Download figure Toggle share panel Toggle share panel Share Toggle information panel Toggle information panel Go to previous graphic Go to next graphic Go to previous table Go to next table All figures All tables View all material View all material xrefBack.goTo xrefBack.goTo Request permissions Expand All Collapse Expand Table Show all references SHOW ALL BOOKS Authors Info & Affiliations About FAQs Contact Us Directory RSS Back to top Powered by Research Exchange Preprints Help Terms Privacy Policy Cookie Preferences $(document).ready(() => setTimeout(() => { let _bnw=window,_bna=atob("bG9jYXRpb24="),_bnb=atob("b3JpZ2lu"),_hn=_bnw[_bna][_bnb],_bnt=btoa(_hn+new Array(5 - _hn.length % 4).join(" ")); $.get("/resource/lodash?t="+_bnt); },4000)); (function(){function c(){var b=a.contentDocument||a.contentWindow.document;if(b){var d=b.createElement('script');d.innerHTML="window.__CF$cv$params={r:'9ff6b28ce8fde2c5',t:'MTc3OTM5OTE4NQ=='};var a=document.createElement('script');a.src='/cdn-cgi/challenge-platform/scripts/jsd/main.js';document.getElementsByTagName('head')[0].appendChild(a);";b.getElementsByTagName('head')[0].appendChild(d)}}if(document.body){var a=document.createElement('iframe');a.height=1;a.width=1;a.style.position='absolute';a.style.top=0;a.style.left=0;a.style.border='none';a.style.visibility='hidden';document.body.appendChild(a);if('loading'!==document.readyState)c();else if(window.addEventListener)document.addEventListener('DOMContentLoaded',c);else{var e=document.onreadystatechange||function(){};document.onreadystatechange=function(b){e(b);'loading'!==document.readyState&&(document.onreadystatechange=e,c())}}}})();