Method for identifying food range of megachilid bees based on multi-technology integration

Abstract

This invention discloses a method for identifying the dietary range of reed wasps based on multi-technology integration. This method addresses the differences in nutritional requirements between adult and larval stages of reed wasps by integrating four technologies: field observation of flowers, microscopic identification of pollen on the wasp's body surface, microscopic identification of mixed pollen masses in nest cells, whole-genome sequencing of mixed pollen masses in nest cells, and bioinformatics analysis. Through standardized sample collection, extraction, identification, and sequencing analysis procedures, combined with cross-validation of multiple technologies, it achieves accurate identification of the dietary range of reed wasps throughout their entire life cycle. Simultaneously, it clarifies the key parameters and judgment criteria for each technical step, constructs a reference sequence library specific to tropical island plants, and verifies the high consistency of results among different technologies using the Kappa coefficient. This invention avoids the limitations of identifying dietary habits using a single technology, achieving full-coverage identification of the adult-larval life cycle of reed wasps with high accuracy and reliable results, providing a standardized technical framework for dietary research on solitary pollinating insects in tropical islands.

Description

Technical Field

[0001] This invention belongs to the field of insect diet identification technology, specifically involving a method for identifying the diet range of reed wasps based on multi-technology integration, which is particularly suitable for the study of the diet of adults and larvae of solitary pollinating insects in tropical islands throughout their entire life cycle. Background Technology

[0002] genus Reed Bee ( Ceratina Insects are an important group within the family Apidae of the order Hymenoptera, widely distributed globally and comprising over 300 species. These bees are small in size, often with a metallic sheen, and have received widespread attention from ecologists due to their crucial role in plant pollination. Studies have confirmed that reed wasps are major pollinators of many agricultural and horticultural crops, and are also indispensable pollination media in natural ecosystems. Their importance is particularly prominent in tropical island ecosystems. The Taiwan reed wasp (… Ceratina taiwanensis ), Green Reed Bee ( Ceratina smaragdula ) and toothed tibiae bee ( Ceratina dentipes Species such as [list of species would be inserted here] are key pollinators in these regions. Due to the isolation and fragility of island ecosystems, the stability of plant pollination networks highly depends on a small number of highly adaptable pollinator groups. As native pollinators, *Pterocarya spp.* (also known as spp.) have formed a close co-evolutionary relationship with island plants and play an irreplaceable role in the construction and maintenance of island vegetation ecosystems. Clarifying the dietary range of these *Pterocarya spp.*, i.e., the phylogenetic distribution of the plant species from which they collect pollen, is of crucial theoretical and practical significance for protecting *Pterocarya spp.* populations on tropical islands and for utilizing these native pollinators to restore and maintain the stability of island vegetation ecosystems.

[0003] Traditionally, the study of bee pollination diets has relied primarily on field observation and pollen morphology identification. Field observation infers a bee's diet by directly recording the plant species it visits. This method is intuitive but time-consuming and labor-intensive, and it struggles to comprehensively reflect the pollen-collecting range of bees throughout their life cycle, especially when bees visit flowers in the difficult-to-observe canopy or outside of survey periods. Pollen morphology identification (microscopic identification) involves collecting pollen carried on the bee's body or pollen masses from the hive and identifying species based on the morphological characteristics of the pollen grains under a microscope. However, traditional microscopic identification techniques have significant limitations: they heavily rely on the experience of specialized technicians, the identification process is time-consuming and labor-intensive, and they typically only identify pollen at the family or genus level, failing to achieve species-level resolution. For morphologically similar pollen grains, the risk of misidentification and omission is high, making it difficult to comprehensively and accurately reveal the true dietary range of pollinators.

[0004] In recent years, with the rapid development of molecular biology techniques, DNA macrobarcoding technology has been widely applied in the field of pollen species identification. This technology extracts total DNA from mixed pollen samples, amplifies specific DNA barcode fragments (such as ITS2 and rbcL) using universal primers, and then performs high-throughput sequencing and bioinformatics analysis. This allows for the simultaneous identification of all plant species contained in the sample, efficiently and accurately revealing pollen collection patterns in bees, identifying plant species that are difficult to detect using traditional methods, and greatly enhancing our understanding of the complexity and diversity of pollinator diets.

[0005] However, DNA metabarcoding technology still suffers from inherent limitations such as primer bias and PCR amplification deviation, making it difficult to meet the demands of accurate identification. Therefore, whole-genome sequencing technology is gradually being applied to pollen species identification. This technology eliminates the need for PCR amplification, directly sequencing the sample and effectively avoiding the biases introduced by PCR amplification, further improving identification accuracy and reliability. However, this technology is still affected by factors such as incomplete reference databases, resulting in some degree of bias in the identification results. Therefore, organically combining traditional methods with modern molecular techniques to achieve complementary advantages is an inevitable trend for accurately determining the dietary range of pollinators.

[0006] Currently, systematic research on the diet of wasps in tropical island regions is lacking. Most existing studies focus on pollination within agricultural ecosystems, while a comprehensive methodological system and technical solutions are lacking regarding the plant resource utilization phylogenetics of wasps in the unique habitats of islands. Single technical approaches are insufficient to overcome their inherent limitations and cannot meet the need for a comprehensive and accurate analysis of the wasp's diet range. Therefore, there is an urgent need to establish an analytical method that integrates multiple technical means, combining field ecological observation with precise laboratory identification, to achieve a three-dimensional, high-resolution identification covering the entire developmental stage of the wasp's diet range. This would provide scientific evidence and technical support for the protection of wasp populations in tropical islands and for the restoration and stability of vegetation ecosystems. Summary of the Invention

[0007] To address the technical problems of existing methods for studying the diet of pollinating insects, such as the inability to distinguish developmental stages, the difficulty in tracing larval diets, and the lack of research on the entire life history diet of tropical island wasps, this invention provides a method for identifying the entire life history diet of wasps based on developmental stage differentiation. This method, considering the differences in nutritional needs between adult and larval stages, divides field flower observation, microscopic identification of pollen on the body surface, microscopic identification of mixed pollen masses in nest cells, whole-genome sequencing, and bioinformatics analysis into two major identification modules: adult and larval stages. Differential cross-validation standards are established to achieve stage-specific diet identification. By clarifying key operational parameters, constructing a plant reference sequence library for the study area, and introducing the Kappa coefficient consistency test, a reproducible and standardized technical framework for studying the entire life history diet of wasps is established.

[0008] To achieve the above objectives, the technical solution adopted by the present invention is as follows:

[0009] A method for identifying the dietary range of reed wasps based on multi-technology integration, taking into account the differences in nutritional requirements between adult and larval stages, divides the identification method into an adult stage dietary identification module and a larval stage dietary identification module, and sets differentiated cross-validation standards. Through modular integration and stratified validation, it achieves stage-specific dietary identification, including the following steps:

[0010] (1) Construction of the adult feeding behavior identification module:

[0011] S1. Field observation of flower visits: Collect individual bees visiting flowers in the study area, record information on the flowers visited, and stratify and screen samples according to the types of flowers visited.

[0012] S2. Microscopic identification of pollen on the body surface: Pollen on the body surface of the selected individual wasps was extracted by ultrasonic-assisted centrifugation. After slide preparation and staining, it was observed under an optical microscope to identify pollen species and count their relative proportions.

[0013] S3. Cross-validation of adult stage: The consistency of the above field flower observation results with the microscopic identification results of pollen on the body surface is tested. The Kappa coefficient is required to be ≥0.85 to determine the range of adult feeding habits.

[0014] (2) Construction of the larval feeding behavior identification module:

[0015] S4. Microscopic identification of mixed pollen masses from nest cells: Mixed pollen masses from nest cells of the bee were collected, pretreated, dispersed and purified, then prepared and stained. The pollen species were identified and their relative proportions were statistically analyzed by microscopic observation.

[0016] S5. Whole-genome sequencing and bioinformatics analysis of mixed pollen masses in nests: Genomic DNA was extracted from mixed pollen masses in nests, a sequencing library was constructed and sequenced, and the sequencing data were compared with the constructed plant reference sequence library of the study area after quality control to identify the plant species of pollen source and to count their relative proportions.

[0017] S6. Cross-validation during the larval stage: The consistency of the above microscopic identification results with the whole genome sequencing results is checked. The Kappa coefficient is required to be ≥0.65 to determine the feeding range during the larval stage.

[0018] (3) Integration of diet throughout life history: Combine the diet species of the adult and larval stages verified in steps (1) and (2), exclude duplicate species, and construct the diet range of the entire life history of the reed wasp.

[0019] Furthermore, the field flower observation described in step S1 includes:

[0020] Collection Records: Collect individual flower-visiting bees in the study area and record the collection location, time, and flower-visiting plant species;

[0021] Stratified sampling: Stratify according to the species of flowering plants, and randomly select 1 to 3 individual bees of each species as the analysis sample. If only 1 specimen is collected for a certain plant, it is directly used as the analysis sample.

[0022] Pairing and screening: Fifteen individual bees were selected as paired samples for microscopic identification of pollen on their body surface, to verify consistency with the results of field flower observation.

[0023] Furthermore, the microscopic identification of pollen on the body surface in step S2 includes: extracting pollen on the body surface using ultrasonic-assisted centrifugation, with ultrasonic power of 40kHz, temperature of 25℃, time of 15min, centrifugation of 5000r / min for 5min, and repeated extraction 2-3 times; after staining the pollen with 1% eosin Y ethanol solution, slides are prepared, and three replicate slides are prepared for each sample. Three fields of view are randomly selected from each slide for observation and identification under a 40× objective lens, and the relative proportion of each pollen species is statistically analyzed.

[0024] Furthermore, the microscopic identification of mixed pollen masses in nest chambers in step S4 includes: softening the pollen masses by soaking in 75% ethanol, ultrasonically dispersing, centrifuging and cleaning for purification, staining with eosin Y to prepare slides, preparing 3 replicate slides for each sample, randomly selecting 3 fields of view on each slide for observation and identification under a 40× objective lens, and statistically analyzing the relative proportion of each pollen species.

[0025] Furthermore, step S4 also includes the following steps: selecting 9 mixed pollen mass samples from nest chambers as paired samples for whole genome sequencing and bioinformatics analysis, and using them to verify consistency with the microscopic identification results of the mixed pollen mass samples from nest chambers.

[0026] Furthermore, the extraction of genomic DNA from mixed pollen cells in step S5 is performed using the CTAB method, which includes: grinding the tissue into powder in liquid nitrogen and transferring it to a centrifuge tube; adding CTAB extraction buffer preheated to 65°C and lysing it in a 65°C water bath; adding an equal volume of chloroform-isoamyl alcohol mixture at a volume ratio of 24:1, vortexing and centrifuging, and collecting the upper aqueous phase; adding 0.6 to 1 volume of isopropanol to precipitate the DNA, centrifuging and discarding the supernatant; washing the precipitate twice with 70% ethanol, centrifuging and discarding the ethanol; drying and dissolving it in TE buffer or ultrapure water, and storing it at -20°C for later use.

[0027] Furthermore, the construction of the sequencing library and sequencing described in step S5 includes: breaking the DNA to 350bp using a Covaris fragmenter, performing end repair, adding A tails, ligating Illumina adapters, and screening, followed by high-fidelity PCR amplification and enrichment, and calibrating the library concentration to 1.5nM; performing PE150 paired-end sequencing using the Illumina Novaseq platform, with each sample having a clean data percentage ≥85%.

[0028] Furthermore, the sequencing data quality control standards in step S5 are as follows: data quality is assessed using FastQC v0.11.9, and Trimmomatic v0.39 software is used for filtering to remove reads containing adapter sequences, low-quality reads with a base quality value Q≤20 accounting for more than 50%, and short reads with a length of less than 50bp; after quality control, the clean data ratio is required to be ≥85%, the GC content to be 35%~55%, and the adapter contamination rate to be <0.5%.

[0029] Furthermore, the method for constructing the plant reference sequence library of the study area in step S5 is as follows: integrate the plant list of the study area, download the chloroplast sequences of the target plants from the NCBI database, and construct a chloroplast reference sequence library without redundancy by quality control and clustering and deduplication using CD-HIT v4.8.1 software; perform sequence alignment using Blasten v2.13.0 software, and remove trace groups with a read ratio of <0.01%; the constructed reference sequence library can realize species-level identification of pollen and can effectively detect low-abundance rare pollen species with a ratio of <0.05%.

[0030] Furthermore, the cross-validation of the adult stage in step (1) and the cross-validation of the larval stage in step (2) both adopt the species-level presence or absence matching method for Kappa consistency test: taking a single sample as an independent statistical unit, the results of different technologies to species identification are subjected to binary classification pairing analysis.

[0031] This invention addresses the differences in nutritional requirements between adult and larval stages of the wasp by integrating four techniques: field observation of flowers, microscopic identification of pollen on the body surface, microscopic identification of mixed pollen masses in nest cells, whole-genome sequencing of mixed pollen masses in nest cells, and bioinformatics analysis. Field observation of flowers and microscopic identification of pollen on the body surface are used to target the dietary habits of adults, while microscopic identification of mixed pollen masses in nest cells, combined with whole-genome sequencing and bioinformatics analysis, are used to target the dietary habits of larvae. Through cross-validation of multiple techniques and statistical consistency analysis, the dietary range of the entire life history of the wasp can be identified. The core analytical steps are performed in the following order, and the parameters of each step are standardized and reproducible:

[0032] 1. Observing flowers in the wild

[0033] The flower-visiting behavior of reed wasps was observed at different time points in the study area. Individual reed wasps that were conducting flower-visiting behavior were collected using insect nets and placed in 75% alcohol specimen bottles. The collection location, collection time, and species of the flower-visiting plant were labeled. Stratified sampling was carried out according to the species of flower-visiting plant. 1 to 3 individuals of each flower-visiting plant were randomly selected as samples for subsequent analysis. If only 1 specimen of a certain plant was collected, it was directly used as the analysis sample to ensure the representativeness of the samples.

[0034] Fifteen individual wasps collected from the wild were selected as paired samples for microscopic identification of pollen on their body surface. These samples were used to verify the consistency with the results of field flower visits, providing basic ecological information for determining the feeding range of adult wasps.

[0035] 2. Microscopic identification of pollen on the body surface

[0036] Pollen was extracted from the surface of individual wasps using ultrasonic-assisted centrifugation. The specific procedure was as follows: individual wasps were placed in sterile centrifuge tubes, sterile deionized water was added, and the tubes were ultrasonically vibrated at 40 kHz and 25 °C for 15 min. Then, the tubes were centrifuged at 5000 r / min for 5 min. The supernatant was discarded, and the process of adding water, ultrasonication, and centrifugation was repeated 2-3 times to ensure complete pollen extraction. 0.2 mL of 1% eosin Y ethanol solution was added to the pollen precipitate, and the tubes were stained at room temperature for 5 min. 10 μL of the pollen suspension was then used to prepare slides, with three replicates for each sample. Three non-edge fields of view were randomly selected from each slide and observed under a 40× objective lens using an optical microscope. High-resolution cameras were used to record the observations. Pollen species were identified using a pollen morphology database, and the number of different pollen species in each field of view was counted. The relative percentage was calculated as "number of a certain pollen species / total pollen count × 100%".

[0037] This step is paired with the results of field flower observations to determine the immediate feeding range of adult reed wasps and to verify inter-technical consistency.

[0038] 3. Microscopic identification of mixed pollen masses in nest cells

[0039] Using sterile forceps, remove the mixed pollen clumps from the honeycomb cells and place them individually in a sterile centrifuge tube. Add 1 mL of 75% ethanol aqueous solution and soak for 10 min to soften. Use sterile forceps to break up large aggregates. Place the tube in an ultrasonic cleaner (40 kHz, 25℃) and vibrate for 20 min to disperse the pollen clumps into individual particles. Centrifuge at 5000 r / min for 5 min, discard the supernatant, and repeat the ethanol washing-centrifugation operation 2-3 times until the supernatant is clear, completing the pollen purification. Aspirate 100 μL of the purified pollen solution, add 0.2 mL of 1% eosin Y ethanol solution, and stain at room temperature for 5 min. Prepare 10 μL of the solution for slide preparation, and prepare 3 replicate slides for each sample. Randomly select 3 non-edge fields of view on each slide for optical microscopy observation with a 40× objective lens and record the images using a high-definition camera. Identify pollen species based on the pollen morphology database, count the number of different pollen species in each field of view, and calculate the relative proportion as "number of a certain type of pollen / total number of pollen × 100%".

[0040] Nine mixed pollen mass samples from nest chambers were selected as paired samples for whole-genome sequencing and bioinformatics analysis to verify consistency with the results of nest chamber microscopic identification, providing morphological basis for determining the larval feeding range.

[0041] 4. Whole-genome sequencing and bioinformatics analysis of mixed pollen masses in nests

[0042] (1) DNA extraction and library construction: Genomic DNA was extracted from mixed pollen masses in nests using the CTAB method. The DNA was fragmented into 350bp target fragments using a Covaris fragment disruptor. After end repair, A-tailing at the 3' end, ligation with Illumina sequencing adapters, and fragment screening, the library was amplified and enriched by high-fidelity PCR. The library concentration was calibrated to 1.5nM using a fluorescence quantitative PCR instrument.

[0043] (2) WGS sequencing: Sequencing was performed using the Illumina Novaseq platform and PE150 paired-end sequencing mode to ensure that each sample obtained sufficient sequence data and that the clean data ratio was ≥85%;

[0044] (3) Data quality control: FastQC v0.11.9 software was used to evaluate the quality of raw sequencing data. Trimmomatic v0.39 software was used to filter and remove reads containing adapter sequences, low-quality reads with a base quality value Q≤20 accounting for more than 50%, and short reads with a length of less than 50bp. The clean data ratio was required to be ≥85%, GC content 35%~55%, and adapter contamination rate <0.5%.

[0045] (4) Construction of reference sequence library: The plant list of the study area was integrated, and the chloroplast sequences of the target plants were downloaded from the NCBI database. After quality control, the CD-HIT v4.8.1 software was used to cluster and remove duplicates to construct a chloroplast sequence reference library without redundancy, which is suitable for the accurate identification of pollen species in the study area.

[0046] (5) Sequence alignment and species identification: Blasten v2.13.0 software was used to perform homology comparison between the clean data after quality control and the reference sequence library. The number of reads in the sequence for each plant species was counted, the relative proportion was calculated, and trace groups with a proportion of <0.01% were removed (considered as sequencing error or environmental pollution). This technology can accurately detect rare pollen species with a proportion of <0.05%, realize the species-level accurate identification and quantitative analysis of pollen-derived plants, and make up for the limitations of microscopic identification.

[0047] 5. Cross-validation of multiple technologies and determination of dietary range

[0048] The identification results from flower observation, microscopic identification of pollen on the body surface, microscopic identification of mixed pollen masses in nest cells, whole-genome sequencing of mixed pollen masses in nest cells, and bioinformatics analysis were subjected to developmental stage-targeted cross-validation, following the calibration principles:

[0049] ① Only identification results at the species level are included, while merged groups at the genus / family level and morphotypes with unclear classifications are excluded to ensure the accuracy of the identification results;

[0050] ② Cross-validation of field flower observation and microscopic identification of pollen on the body surface, combined with non-repeating species, determined the dietary range of adult reed wasps; cross-validation of microscopic identification of mixed pollen masses in nest cells and whole genome sequencing and bioinformatics analysis, combined with non-repeating species, determined the dietary range of reed wasps during the larval stage.

[0051] ③ Combine the feeding species of the adult and larval stages, count the range of plant families, genera and species covered by the diet of the reed wasp, and determine the diet range of the reed wasp throughout its entire life cycle;

[0052] ④ The consistency of results between techniques was verified by the Kappa coefficient. The Kappa coefficient of field flower observation and microscopic identification of pollen on the body surface was ≥0.85 (almost completely consistent), and the Kappa coefficient of microscopic identification of mixed pollen masses in nest cells and whole genome sequencing and bioinformatics analysis was ≥0.65 (highly consistent), ensuring the reliability of the identification results.

[0053] The present invention has the following beneficial effects:

[0054] 1. Achieving Identification of the Entire Life History of Feeding Habits: This invention does not simply integrate existing technologies, but rather designs a unique combination of technologies tailored to the nutritional differences between adult and larval stages of the reed wasp. Field observation of flowers combined with microscopic identification of pollen on the body surface analyzes the adult's feeding habits, while microscopic identification of mixed pollen masses in the nest chamber combined with whole-genome sequencing and bioinformatics analysis analyzes the larval's feeding habits. This achieves a systematic identification of the entire life history of the reed wasp, from adult to larval stages, filling a gap in research on the specific feeding habits of reed wasps at different developmental stages on tropical islands.

[0055] 2. Leveraging the complementary advantages of multiple technologies to avoid the limitations of single technologies and establish quantitative verification standards: Integrating three major categories of technologies: field flower observation, pollen morphology identification, and whole-genome sequencing and bioinformatics analysis. Among these, whole-genome sequencing avoids PCR amplification bias and can detect trace pollen groups accounting for <0.05%, significantly improving detection sensitivity. Through cross-validation of multiple technologies, quantitative judgment standards are established, reducing the problems of missed detections and false positives associated with single technologies, while also improving the reliability and reproducibility of identification results.

[0056] 3. Establishing a dedicated plant reference sequence library tailored to the unique habitats of islands and the biological characteristics of solitary wasps: Based on the plant community composition characteristics of the unique habitats of tropical islands and the behavioral characteristics of solitary wasps such as nest building and pollen storage, a plant barcode reference sequence library specific to the study area was constructed. This database covers the genome or barcode sequences of local potential nectar and pollen source plants, significantly improving the accuracy and regional adaptability of pollen species identification, and overcoming the shortcomings of general databases in identifying endemic species.

[0057] 4. Standardized operation and high repeatability: The key parameters, judgment criteria and data analysis methods of each analysis step are clearly defined, and a standardized identification system for the entire life history and diet range of reed wasps is constructed, which can be directly repeated by those skilled in the art; at the same time, a plant reference sequence library specific to the study area is constructed, which improves the accuracy and adaptability of pollen species identification. Attached Figure Description

[0058] Figure 1 This is a flowchart illustrating the method for identifying the dietary range of reed wasps based on multi-technology integration as described in this invention. Detailed Implementation

[0059] The following embodiments are further illustrations of the present invention, but not limitations thereof.

[0060] Example 1: Field Flower Observation

[0061] Observations were conducted at different time points at the Wenchang Island Plant Base in Hainan Province, China, from August 30 to September 13, 2023 (temperature 26-31℃, weather mainly thunderstorms and cloudy). Individuals of reed wasps exhibiting flower-visiting behavior were collected using insect nets and placed in specimen bottles containing 75% alcohol. The collection location, time, flower-visiting plant species, and collector information were labeled.

[0062] A total of 66 wasp specimens were collected, involving 18 flowering plants and 11 families. Stratified sampling was performed according to the flowering plant species, and 1 to 3 wasp individuals were randomly selected from each plant species. Finally, 15 wasp individuals were selected as paired samples for microscopic identification of pollen on their body surface. This was used to verify the consistency of adult feeding habits with the results of field observations. The individual numbers were in the format of "plant species abbreviation-individual serial number".

[0063] Table 1 Results of field flower observation

[0064] Example 2: Microscopic identification of pollen on the body surface

[0065] 1. Pollen extraction: The 15 individual bees selected in Example 1 were placed in sterile centrifuge tubes, 3 mL of sterile deionized water was added, and the tubes were placed in an ultrasonic cleaner (40 kHz, 25 °C) and vibrated for 15 min; centrifuged at 5000 r / min for 5 min and the supernatant was discarded; the "add water-ultrasound-centrifugation" operation was repeated twice to ensure that the pollen was fully extracted.

[0066] 2. Slide preparation and staining: Add 0.2 mL of 1% eosin Y ethanol solution to the pollen precipitate, mix well, and stain at room temperature for 5 min; take 10 μL of pollen suspension to prepare slides, prepare 3 replicate slides for each sample, and let stand at room temperature for 10 min to prepare permanent slides;

[0067] 3. Microscopic identification: Place the slide under an optical microscope and observe with a 40× objective lens. Randomly select three non-edge fields of view for each slide and take microscopic photographs. Identify pollen species based on "Pollen Morphology of Tropical and Subtropical Angiosperms in China" and a self-built pollen morphology database. Count the number of different pollen species in each field of view and calculate the relative proportion.

[0068] 4. Verification of consistency of adult stage techniques: The results of microscopic identification of pollen on the body surface of 15 samples in this example were compared with the results of field observation of 15 reed wasps in Example 1 using Kappa consistency analysis. The general judgment standard in the field of biological identification (Kappa value of 0.81~1.00 is almost completely consistent, and 0.61~0.80 is highly consistent) was adopted. The results showed that the Kappa coefficient was 0.889 and the chi-square test showed P<0.001. This indicates that the results of field observation and microscopic identification of pollen on the body surface are almost completely consistent in identifying the diet of reed wasps in the adult stage, which verifies the reliability of the two techniques and provides dual support for determining the range of diet in the adult stage.

[0069] Table 2. Microscopic identification results of pollen on the body surface

[0070] Example 3: Microscopic identification and analysis of mixed pollen masses in nest cells

[0071] 1. Sample pretreatment: Select mixed pollen masses from cells 1 to 9 of the Yongshu Reef Bee nest, remove them completely with sterile forceps and place them in a sterile centrifuge tube; add 1 mL of 75% ethanol aqueous solution, soak for 10 min to soften, and break up large aggregates with sterile forceps;

[0072] 2. Dispersion and purification: Place the centrifuge tube in an ultrasonic cleaner (40kHz, 25℃) and vibrate for 20 minutes to disperse the pollen clusters into individual particles; centrifuge at 5000r / min for 5 minutes and discard the supernatant; repeat the ethanol washing-centrifugation operation twice until the supernatant is clear;

[0073] 3. Slide preparation and staining: Mix the pollen solution, extract 100 μL and transfer it to a new centrifuge tube, add 0.2 mL of 1% eosin Y ethanol solution, and stain at room temperature for 5 min; extract 10 μL to prepare slides, and prepare 3 replicate slides for each sample. Let them stand at room temperature for 10 min to make permanent slides.

[0074] 4. Microscopic Identification: Under a microscope, three non-marginal fields of view were selected for each slide, with at least 200 pollen grains counted in each field of view. Images were captured using a high-resolution camera. Pollen species were identified based on "Pollen Morphology of Tropical and Subtropical Angiosperms in China" and a self-built pollen morphology database. The number of different pollen grains in each field of view was counted, and their relative proportions were calculated. Due to the high similarity of pollen morphology, some pollen grains can only be identified to the family or genus level through microscopic identification. In this invention, the precise species-level identification of dietary range is based on the results of whole-genome sequencing and bioinformatics analysis of mixed pollen masses in nests.

[0075] Table 3. Microscopic identification results of mixed pollen masses from nest cells.

[0076] Example 4: Whole-genome sequencing and bioinformatics analysis of mixed pollen masses in nests

[0077] 1. DNA extraction and library construction: The remaining pollen from the mixed pollen masses of cells 1-9 in Example 3 was selected, and genomic DNA was extracted using the CTAB method. The DNA was broken into 350bp target fragments, and after end repair, 3' end A tail addition, ligation of sequencing adapters, and fragment screening, the enriched library was amplified by high-fidelity PCR and the library concentration was calibrated to 1.5nM.

[0078] 2. Whole genome sequencing: The standardized library was loaded onto the Illumina Novaseq platform, and PE150 paired-end sequencing was performed;

[0079] 3. Data quality control: FastQC v0.11.9 was used to assess data quality, and Trimmomatic v0.39 was used to filter low-quality data. The percentage of clean data in this sequencing was ≥88%, the GC content was about 50%, and the adapter contamination rate was <0.5%, which met the quality control standards. Each sample yielded no less than 8Gb of high-quality data.

[0080] 4. Construction of reference sequence library: The plant list of Yongshu Reef was integrated, 232 target plants were screened, their chloroplast sequences were downloaded from NCBI, and after quality control, CD-HIT v4.8.1 software was used for clustering and deduplication to construct a reference sequence library containing 516 non-redundant sequences.

[0081] 5. Sequence alignment and species identification: Blasten v2.13.0 software was used to align the clean data with the reference sequence library (parameters: E-value=1e-10, identity=99%, query coverage=90%), calculate the relative proportion of each plant species, and remove trace groups with a proportion <0.01% as interference.

[0082] Results: A total of 33 plant pollen species were detected, covering 31 genera and 14 families, and all were accurately identified to the species level. Solanaceae's Solanum nigrum and Physalis alkekengi, and Lamiaceae's Vitex trifolia were the core dominant species. At the same time, rare species with low proportions, such as Eriocaulon buergerianum and Eriocaulon buergerianum, were also detected, which made up for the limitations of morphological identification.

[0083] 6. Verification of consistency of larval stage techniques: The results of microscopic identification of mixed pollen masses in nest chambers in Example 3 and the results of whole-genome sequencing and bioinformatics analysis in Example 4 were subjected to a species-level pairing consistency test. Species identified by both methods were used as matching objects, and statistical analysis was performed using the Kappa test. The results showed that Kappa = 0.691 (P < 0.001), indicating a high level of consistency between the microscopic identification of pollen masses in nest chambers and the results of whole-genome sequencing and bioinformatics analysis, thus verifying the reliability and stability of the multi-technology combination identification results of this invention.

[0084] Table 4. Results of whole-genome sequencing and bioinformatics analysis of mixed pollen cells in the nest.

[0085] Example 5: Cross-validation of multiple technologies and determination of dietary range

[0086] By integrating the results of field flower observation in Example 1, microscopic identification of pollen on body surface in Example 2, microscopic identification of pollen in nest chamber in Example 3, and whole genome sequencing and bioinformatics analysis of pollen in nest chamber in Example 4, the dietary range of adult, larval, and entire life history of the reed wasp was determined through cross-validation and Kappa consistency test.

[0087] 1. Verification of Adult Feeding Habits: In Example 1, flower-visiting observations were conducted on 15 individual wasps, recording a total of 15 plant species visited. In Example 2, pollen samples from the same batch of individuals were identified, revealing 16 pollen species, including 14 species, 1 genus, and 1 family. Paired comparison of the two sets of results showed a Kappa ratio of 0.889 (P < 0.001), indicating near-identical consistency. By excluding non-repeating species and removing undetermined family and genus groups, the feeding range of adult wasps was determined.

[0088] 2. Verification of Larval Diet: In Example 3, microscopic identification of pollen in the nest chambers detected 18 pollen species, including 13 species, 4 families, 1 genus, and 1 species with unknown morphology. In Example 4, whole-genome sequencing and bioinformatics analysis of pollen in the nest chambers identified 33 plant species, all at the species level, including several micro-species with relative abundance below 0.05%. The two sets of results showed high agreement on dominant species, with Kappa = 0.691 (P < 0.001), reaching a high level of consistency. By excluding non-repeating species and removing family, genus, and unknown morphological groups, the dietary range of the reed wasp larvae was determined.

[0089] 3. Determination of the entire life history diet range: By merging the species-level identification results of the adult and larval stages and eliminating duplicate groups, the species-level identification results of the entire life history diet range of the adult-larval reed wasp were finally obtained (i.e., the diet of the Taiwan reed wasp covers 46 species of plants in 43 genera and 19 families). The entire life history diet lineage of the adult-larval reed wasp was constructed, realizing a specific, comprehensive and accurate analysis of the diet range at each developmental stage.

[0090] Table 5. Results of Dietary Range Throughout the Life History of Wasps

[0091] The above are merely preferred embodiments of the present invention. It should be noted that the above preferred embodiments should not be considered as limitations on the present invention, and the scope of protection of the present invention should be determined by the scope defined in the claims. For those skilled in the art, several improvements and modifications can be made without departing from the spirit and scope of the present invention, and these improvements and modifications should also be considered within the scope of protection of the present invention.

Claims

1. A method for identifying the dietary range of reed wasps based on multi-technology integration, characterized in that, To address the differences in nutritional requirements between adult and larval stages of the wasp, the identification method is divided into an adult feeding identification module and a larval feeding identification module. Differentiated cross-validation standards are established, and stage-specific feeding identification is achieved through modular integration and stratified validation. The process includes the following steps: (1) Construction of the adult feeding behavior identification module: S1. Field observation of flower visits: Collect individual bees visiting flowers in the study area, record information on the flowers visited, and stratify and screen samples according to the types of flowers visited. S2. Microscopic identification of pollen on the body surface: Pollen on the body surface of the selected individual wasps was extracted by ultrasonic-assisted centrifugation. After slide preparation and staining, it was observed under an optical microscope to identify pollen species and count their relative proportions. S3. Cross-validation of adult stage: The consistency of the above field flower observation results with the microscopic identification results of pollen on the body surface is tested. The Kappa coefficient is required to be ≥0.85 to determine the range of adult feeding habits. (2) Construction of the larval feeding behavior identification module: S4. Microscopic identification of mixed pollen masses from nest cells: Mixed pollen masses from nest cells of the bee were collected, pretreated, dispersed and purified, then prepared and stained. The pollen species were identified and their relative proportions were statistically analyzed by microscopic observation. S5. Whole-genome sequencing and bioinformatics analysis of mixed pollen masses in nests: Genomic DNA was extracted from mixed pollen masses in nests, a sequencing library was constructed and sequenced, and the sequencing data were compared with the constructed plant reference sequence library of the study area after quality control to identify the plant species of pollen source and to count their relative proportions. S6. Cross-validation during the larval stage: The consistency of the above microscopic identification results with the whole genome sequencing results is checked. The Kappa coefficient is required to be ≥0.65 to determine the feeding range during the larval stage. (3) Integration of diet throughout life history: Combine the diet species of the adult and larval stages verified in steps (1) and (2), exclude duplicate species, and construct the diet range of the entire life history of the reed wasp.

2. The method according to claim 1, characterized in that, The field flower observation described in step S1 includes: Collection Records: Collect individual flower-visiting bees in the study area and record the collection location, time, and flower-visiting plant species; Stratified sampling: Stratify according to the species of flowering plants, and randomly select 1 to 3 individual bees of each species as the analysis sample. If only 1 specimen is collected for a certain plant, it is directly used as the analysis sample. Pairing and screening: Fifteen individual bees were selected as paired samples for microscopic identification of pollen on their body surface, to verify consistency with the results of field flower observation.

3. The method according to claim 1, characterized in that, The microscopic identification of pollen on the body surface in step S2 includes: extracting pollen on the body surface using ultrasonic-assisted centrifugation, with ultrasonic power of 40kHz, temperature of 25℃, time of 15min, centrifugation of 5000r / min for 5min, and repeated extraction 2-3 times; after staining the pollen with 1% eosin Y ethanol solution, slides are prepared, and three replicate slides are prepared for each sample. Three fields of view are randomly selected from each slide for observation and identification under a 40× objective lens, and the relative proportion of each pollen species is statistically analyzed.

4. The method according to claim 1, characterized in that, The microscopic identification of mixed pollen masses in nest chambers in step S4 includes: softening the pollen masses by soaking in 75% ethanol, ultrasonically dispersing, centrifuging and cleaning for purification, staining with eosin Y to prepare slides, preparing 3 replicate slides for each sample, randomly selecting 3 fields of view on each slide for observation and identification under a 40× objective lens, and statistically analyzing the relative proportion of each pollen species.

5. The method according to claim 1 or 4, characterized in that, Nine mixed pollen mass samples from nest chambers were selected as paired samples for whole-genome sequencing and bioinformatics analysis to verify consistency with the microscopic identification results of the mixed pollen mass samples from nest chambers.

6. The method according to claim 1, characterized in that, The extraction of genomic DNA from mixed pollen cells in step S5 is performed using the CTAB method, which includes: grinding the tissue into powder in liquid nitrogen and transferring it to a centrifuge tube; adding CTAB extraction buffer preheated to 65°C and lysing it in a 65°C water bath; adding an equal volume of chloroform-isoamyl alcohol mixture at a volume ratio of 24:1, vortexing and centrifuging, and collecting the upper aqueous phase; adding 0.6 to 1 volume of isopropanol to precipitate the DNA, centrifuging and discarding the supernatant; washing the precipitate twice with 70% ethanol, centrifuging and discarding the ethanol; drying and dissolving it in TE buffer or ultrapure water, and storing it at -20°C for later use.

7. The method according to claim 1, characterized in that, The construction of the sequencing library and sequencing described in step S5 includes: breaking DNA to 350bp using a Covaris fragmenter, performing end repair, adding A tails, ligating Illumina adapters, and screening, followed by high-fidelity PCR amplification and enrichment, and calibrating the library concentration to 1.5nM; performing PE150 paired-end sequencing using the Illumina Novaseq platform, with each sample having a clean data percentage ≥85%.

8. The method according to claim 1, characterized in that, The quality control standards for sequencing data in step S5 are as follows: data quality is assessed using FastQC v0.11.9, and Trimmomatic v0.39 software is used for filtering to remove reads containing adapter sequences, low-quality reads with a base quality value Q≤20 accounting for more than 50%, and short reads with a length of less than 50bp; after quality control, the clean data ratio is required to be ≥85%, the GC content is 35%~55%, and the adapter contamination rate is <0.5%.

9. The method according to claim 1, characterized in that, The method for constructing the plant reference sequence library of the study area in step S5 is as follows: integrate the plant list of the study area, download the chloroplast sequences of the target plants from the NCBI database, and construct a chloroplast reference sequence library without redundancy by quality control and clustering and deduplication using CD-HIT v4.8.1 software; perform sequence alignment using Blasten v2.13.0 software, and remove trace groups with a read ratio of <0.01% in the alignment; the constructed reference sequence library can realize species-level identification of pollen and can effectively detect low-abundance rare pollen species with a ratio of <0.05%.

10. The method according to claim 1, characterized in that, The cross-validation of the adult stage in step (1) and the cross-validation of the larval stage in step (2) both use the species-level presence or absence matching method to perform Kappa consistency test: taking a single sample as an independent statistical unit, the results of different technologies to identify species are subjected to binary classification and pairing analysis.

Images