Rapid Species Identification Research Articles

New Species-Specific Real-Time PCR Assays for Colletotrichum Species Causing Bitter Rot of Apple.

Bitter rot of apple is an economically important worldwide disease caused by different Colletotrichum species, depending on many factors such as climate, geography, other hosts, and crop management practices. Culture, morphology, and single-locus sequencing-based methods for identifying the Colletotrichum species are severely limited in effectiveness, while the multilocus sequence typing methods available for delineating species are costly, time-intensive, and require high expertise. We developed species-specific hydrolysis probe real-time PCR assays for the following nine Colletotrichum species causing bitter rot in the Mid-Atlantic U.S.A.: C. fructicola, C. chrysophilum, C. noveboracense, C. gloeosporioides s.s., C. henanense, C. siamense and C. theobromicola from the C. gloeosporioides species complex, and C. fioriniae and C. nymphaeae from the C. acutatum species complex. After searching 14 gene regions, we designed primers and probes in 5 of them for the nine target species. Four primer-probe set pairs were able to be duplexed. Sensitivity tests showed as little as 0.5 pg DNA were detectable. These real-time PCR assays will provide rapid and reliable identification of these key Colletotrichum species and will be critically important for studies aiming to elucidate their biology, epidemiology, and management on apples as the number one produced and consumed tree fruit in the U.S.A.

Multiplex Detection of Foodborne Pathogens using 3D Nanostructure Swab and Deep Learning-Based Classification of Raman Spectra.

Proactive management of foodborne illness requires routine surveillance of foodborne pathogens, which requires developing simple, rapid, and sensitive detection methods. Here, a strategy is presented that enables the detection of multiple foodborne bacteria using a 3D nanostructure swab and deep learning-based Raman signal classification. The nanostructure swab efficiently captures foodborne pathogens, and the portable Raman instrument directly collects the Raman signals of captured bacteria. a deep learning algorithm has been demonstrated, 1D convolutional neural network with binary labeling, achieves superior performance in classifying individual bacterial species. This methodology has been extended to mixed bacterial populations, maintaining accuracy close to 100%. In addition, the gradient-weighted class activation mapping method is used to provide an investigation of the Raman bands for foodborne pathogens. For practical application, blind tests are conducted on contaminated kitchen utensils and foods. The proposed technique is validated by the successful detection of bacterial species from the contaminated surfaces. The use of a 3D nanostructure swab, portable Raman device, and deep learning-based classification provides a powerful tool for rapid identification (≈5 min) of foodborne bacterial species. The detection strategy shows significant potential for reliable food safety monitoring, making a meaningful contribution to public health and the food industry.

Rapid seafood fraud detection powered by multiple technologies: Food authenticity using DNA-QR codes

Crabs are extremely popular in global seafood market due to their flavour and nutritional content. However, these are frequently contaminated with inedible ones to meet expanding demand and go undetected, putting people's health at risk. In such scenario, genetic testing is the most persuasive approach to authenticate constituent species. Here we bought 11 crabs from a local market at Indian Sundarbans and created barcodes. We for the first time cross-validated our data with other approaches: Barcode-Restriction Fragment Length Polymorphism, Mini-barcoding and Barcode-Melt Curve Analyis along with subsequent data transformation into QR-code and draw a comparative assessment about the efficacy of each approach under various circumstances. We observed that 36% of the examined crabs are edible, whereas 64% are contaminated with inedible ones. We further demonstrated that Bar-RFLP is the most affordable efficient technique in addressing adulteration whereas Bar-MCA is more powerful than sequencing in recognizing very low levels of adulterant species from processed and preserved foods. Thus, we concluded that both Bar-RFLP and Bar-MCA along with the QR-code as data transformation and archival, are effective and rapid species identification tools that can be used instead of traditional sequencing-based approaches requiring extensive computational and data processing infrastructure, saving time and money.

A Multi-Scale Convolutional Neural Network Combined with a Portable Near-Infrared Spectrometer for the Rapid, Non-Destructive Identification of Wood Species

The swift and non-destructive classification of wood species holds crucial significance for the utilization and trade of wood resources. Portable near-infrared (NIR) spectrometers have the potential for rapid and non-destructive wood species identification, and while several studies have explored related methodologies, further research on their practical application is needed. To address this research gap, this study proposes a multi-scale convolutional neural network (CNN) combined with a portable NIR spectrometer (wavelengths range: 908 to 1676 nm) for wood species identification. To enhance the capability of directly extracting robust features from NIR spectral data collected by a portable spectrometer, the Gramian angular field (GAF) method is introduced to transform 1-dimensional (1D) NIR spectral data into 2-dimensional (2D) data matrices. Furthermore, a multi-scale CNN model is utilized for direct feature extraction. The representation by 2D matrices, instead of 1D NIR spectral data, aligns with 2D convolutional operations and enables a more robust extraction of discriminative features. In the experimental phase, eight wood species were identified using the proposed method, alongside commonly used multivariate data analysis and machine learning (ML) methods. The StratifiedGroupKFold dataset partitioning approach and five-fold cross-validation were used. Additionally, nine spectral preprocessing methods were compared, and principal component analysis (PCA) was used for feature extraction in the ML method. Evaluation metrics, such as accuracy, precision, and recall, were adopted to assess the performance of the methods. The proposed multi-scale CNN model, in combination with 2D GAF matrices of the 1D spectral data, yielded the most accurate results with a mean accuracy of 97.34% in the five-fold validation. These findings present a new approach for the construction of a rapid, non-destructive, and automatic wood species identification method using a portable NIR spectrometer.

Rapid identification,fluconazole and micafungin susceptibility testing of Candida species from blood culture by a short incubation method

This study aimed to develop and validate a rapid method for identification by MALDI-TOF system and determination of the susceptibility to Fluconazole and Micafungin by broth microdilution among Candidaspecies causing bloodstream infections. Subcultures from blood culture bottles were incubated for 5 hours (+/- 1h) and used to perform the tests, so that the turnaround time of rapid identification and susceptibility profile was about 5 and 24 hours, respectively. The rapid identification showed agreement of 92.05 %. Regarding the rapid broth microdilution for Fluconazole and Micafungin, the agreement was 97.06 % (p<0.001) and 100 % (p<0.001), and the Kappa coefficient was 0.91 (p<0.001) and 1.0 (p<0.001), respectively. To conclude, both rapid methods showed to be reproducible, inexpensive, easy to perform and time-saving. Thus, these methodologies could be useful to guide and adjust empirical antifungal therapy.

DNA Barcoding of Locusta migratoria manilensis (Orthoptera: Acrididae) Reveals Insights into the Species and Subspecies Differentiation

Abstract Accurate identification and classification of insect species, especially those with significant economic and ecological implications, have historically presented challenges. Migratory locusts, Locusta migratoria manilensis (Meyen) (Orthoptera: Acrididae), are notorious for their destructive impact on crops. Traditional morphological methods often face limitations in distinguishing closely related species and require taxonomic expertise. However, the emergence of DNA barcoding as a powerful tool for species identification has revolutionized the field of entomology. DNA barcoding utilizes a standardized DNA sequence, a molecular barcode, which serves as a distinct genetic signature for rapid and accurate species identification. In this study, DNA barcoding techniques were employed to identify and differentiate the migratory locust subspecies manilensis, in both its solitary and gregarious forms, as well as to determine its phylogenetic relationship with other related species within the Acrididae family. GenBank reference sequences were used to identify the locusts at the molecular subspecies level. Although the COI marker did not exhibit significant differences between the solitary and migratory forms, it was valuable in resolving the identification of L. migratoria subspecies. This lack of significant differences may be attributed to limited genetic variation of COI at the subspecies level and substantial genetic similarities between the solitary and migratory forms, likely stemming from a recent common ancestor. Nonetheless, using COI remains beneficial for subspecies identification in migratory locusts.

Nontuberculous mycobacteria testing and culture positivity in the United States

BackgroundNontuberculous mycobacteria (NTM) are environmental bacteria which may cause chronic lung disease. The prevalence of NTM pulmonary infection and disease has been increasing in the United States and globally. The predominant clinically relevant species of NTM in the United States are Mycobacterium avium complex (MAC) species and Mycobacterium abscessus. With the development of rapid species identification methods for NTM (e.g. PCR probes), more testing for NTM is being conducted through commercial labs, such as Laboratory Corporation of America (Labcorp), which provides deidentified real-time testing data to the Centers for Disease Control (CDC) pursuant to a data sharing agreement. Because NTM lung infections are not reportable in most states, other data sources are key to understanding NTM testing patterns, positivity rates, and species distributions to track infection trends and identify clinical care needs.MethodsWe obtained national Labcorp data for the period January 2019 through mid-April 2022. We subset the data to only respiratory samples sent for Acid Fast Bacilli (AFB) cultures. NTM positive results were defined as those which identified an NTM species and are not Mycobacterium tuberculosis, Mycobacterium bovis, or Mycobacterium gordonae.ResultsOverall, 112,528 respiratory samples were sent for AFB testing during the study period; 26.3% were from the Southeast U.S., identified as HSS Region IV in the Labcorp dataset, and 23.0% were from the Pacific and South Pacific region (Region IX). The culture positive prevalence ranged from 20.2% in the Southeast to 9.2% in the East North Central region (Region V). In the Southeast US, M. abscessus prevalence was 4.0%. For MAC, the highest prevalence was observed in the Mountain region (Region VII) (13.5%) and the lowest proportion was in the East South Central region (7.3%, Region III). Among positive tests, the proportion which was MAC varied from 61.8% to 88.9% and was highest in the Northeast U.S. The proportion of positive samples which were M. abscessus ranged from 3.8% to 19.7% and was highest in the Southeast.ConclusionsThe Southeastern region of the U.S. has the highest rate of culture positivity in Labcorp tests for total NTM and, of all positive tests, the highest proportion of M.abscessus. These estimates may underrepresent the true number of M.abscessus infections because M. absesscus-specific probes are not commercially available and not all NTM testing in the United States is done by Labcorp. Analysis of real-time testing data from commercial laboratories may provide insights into risk factors for NTM culture positivity in ‘hotspot’ areas.

Rapid and accurate identification of bacteria utilizing laser-induced breakdown spectroscopy.

Timely and accurate identification of harmful bacterial species in the environment is paramount for preventing the spread of diseases and ensuring food safety. In this study, laser-induced breakdown spectroscopy technology was utilized, combined with four machine learning methods - KNN, PCA-KNN, RF, and SVM, to conduct classification and identification research on 7 different types of bacteria, adhering to various substrate materials. The experimental results showed that despite the nearly identical elemental composition of these bacteria, differences in the intensity of elemental spectral lines provide crucial information for identification of bacteria. Under conditions of high-purity aluminum substrate, the identification rates of the four modeling methods reached 74.91%, 84.05%, 85.36%, and 96.07%, respectively. In contrast, under graphite substrate conditions, the corresponding identification rates reached 96.87%, 98.11%, 98.93%, and 100%. Graphite is found to be more suitable as a substrate material for bacterial classification, attributed to the fact that more characteristic spectral lines are excited in bacteria under graphite substrate conditions. Additionally, the emission spectral lines of graphite itself are relatively scarce, resulting in less interference with other elemental spectral lines of bacteria. Meanwhile, SVM exhibited the highest precision rate and recall rate, reaching up to 1, making it the most effective classification method in this experiment. This study provides a valuable approach for the rapid and accurate identification of bacterial species based on LIBS, as well as substrate selection, enhancing efficient microbial identification capabilities in fields related to social security and military applications.

Rapid detection of six Oceanobacillus species in Daqu starter using single-cell Raman spectroscopy combined with machine learning.

Many traditional fermented foods and beverages industries around the world request the addition of multi-species starter cultures. However, the microbial community in starter cultures is subject to fluctuations due to their exposure to an open environment during fermentation. A rapid detection approach to identify the microbial composition of starter culture is essential to ensure the quality of the final products. Here, we applied single-cell Raman spectroscopy (SCRS) combined with machine learning to monitor Oceanobacillus species in Daqu starter, which plays crucial roles in the process of Chinese baijiu. First, a total of six Oceanobacillus species (O. caeni, O. kimchii, O. iheyensis, O. sojae, O. oncorhynchi subsp. Oncorhynchi and O. profundus) were detected in 44 Daqu samples by amplicon sequencing and isolated by pure culture. Then, we created a reference database of these Oceanobacillus strains which correlated their taxonomic data and single-cell Raman spectra (SCRS). Based on the SCRS dataset, five machine-learning algorithms were used to classify Oceanobacillus strains, among which support vector machine (SVM) showed the highest rate of accuracy. For validation of SVM-based model, we employed a synthetic microbial community composed of varying proportions of Oceanobacillus species and demonstrated a remarkable accuracy, with a mean error was less than 1% between the predicted result and the expected value. The relative abundance of six different Oceanobacillus species during Daqu fermentation was predicted within 60 min using this method, and the reliability of the method was proved by correlating the Raman spectrum with the amplicon sequencing profiles by partial least squares regression. Our study provides a rapid, non-destructive and label-free approach for rapid identification of Oceanobacillus species in Daqu starter culture, contributing to real-time monitoring of fermentation process and ensuring high-quality products.

Rapid and robust bacterial species identification using hyperspectral microscopy and gram staining techniques.

Gram staining can classify bacterial species into two large groups based on cell wall differences. Our study revealed that within the same gram group (gram-positive or gram-negative), subtle cell wall variations can alter staining outcomes, with the peptidoglycan layer and lipid content significantly influencing this effect. Thus, bacteria within the same group can also be differentiated by their spectra. Using hyperspectral microscopy, we identified six species of intestinal bacteria with 98.1% accuracy. Our study also demonstrated that selecting the right spectral band and background calibration can enhance the model's robustness and facilitate precise identification of varying sample batches. This method is suitable for analyzing bacterial community pathologies.

Recent Advances in Closed-Tube Barcoding for FastFish-ID.

FastFish-ID for rapid and accurate identification of fish species was conceived at Brandeis University based on pioneering work on Closed-Tube Barcoding (Rice et al., Mitochondrial DNA Part A 27(2):1358-1363, 2016; Sirianni et al., Genome 59:1049-1061, 2016). FastFish-ID was subsequently validated and commercialized at Thermagenix, Inc. using a portable device and high-precision PCR (Naaum et al., Food Res Int 141:110035, 2021). The motivation for these efforts was the pressing need for a technology that could be widely used throughout the seafood supply chain to combat IUU Fishing (Helyar et al., PLOS ONE 9, 2014) and overfishing (FAO, State of the World Fisheries and Aquaculture 2018. http://www.fao.org/documents/card/en/c/I9540EN/ , 2018), along with seafood fraud and mislabeling (Watson et al., Fish Fish 17:585-595, 2015). These destructive practices are wasting fish stocks, frustrating attempts to achieve seafood sustainability, endangering oceanic ecosystems, and causing consumers billions of dollars each year (Porterfield et al., Oceana: February, 2022). During the past three Covid19 pandemic years, EcologeniX, LLC has taken over further development and optimization of FastFish-ID. The present chapter provides an overview of the improvements introduced throughout the FastFish-ID process.

Analysis of Whole-Genome facilitates rapid and precise identification of fungal species.

Fungal identification is a cornerstone of fungal research, yet traditional molecular methods struggle with rapid and accurate onsite identification, especially for closely related species. To tackle this challenge, we introduce a universal identification method called Analysis of whole GEnome (AGE). AGE includes two key steps: bioinformatics analysis and experimental practice. Bioinformatics analysis screens candidate target sequences named Targets within the genome of the fungal species and determines specific Targets by comparing them with the genomes of other species. Then, experimental practice using sequencing or non-sequencing technologies would confirm the results of bioinformatics analysis. Accordingly, AGE obtained more than 1,000,000 qualified Targets for each of the 13 fungal species within the phyla Ascomycota and Basidiomycota. Next, the sequencing and genome editing system validated the ultra-specific performance of the specific Targets; especially noteworthy is the first-time demonstration of the identification potential of sequences from unannotated genomic regions. Furthermore, by combining rapid isothermal amplification and phosphorothioate-modified primers with the option of an instrument-free visual fluorescence method, AGE can achieve qualitative species identification within 30 min using a single-tube test. More importantly, AGE holds significant potential for identifying closely related species and differentiating traditional Chinese medicines from their adulterants, especially in the precise detection of contaminants. In summary, AGE opens the door for the development of whole-genome-based fungal species identification while also providing guidance for its application in plant and animal kingdoms.

Recent Applications of FastFish-ID.

FastFish-ID via Closed-Tube barcoding is a portable platform for rapid and accurate identification of fish species that was conceived at Brandeis University, commercialized at Thermagenix, Inc., and further improved at Ecologenix, LLC (see Chap. 17 in this volume). This chapter focuses on the use of FastFish-ID for (1) identification of intraspecies variants, (2) quantitative use of FastFish-ID to measure the decay of fresh fish, and (3) use of FastFish-ID for the identification of dried and processed shark fins.

Fast and on-site animal species identification in processed meat via centrifugal microfluidics and isothermal amplification.

We present a novel centrifugal microfluidic approach to rapidly identify animal species in meat products. The workflow requires a centrifugal cartridge for DNA extraction and for preparation of a recombinant polymerase amplification (RPA) reaction, a programmable centrifuge for processing the cartridge and an isothermal reader to perform the RPA. Liquid reagents are pre-stored on the cartridge and the meat sample can be added directly without any pre-treatment. With this system, we are able to identify six different animal species in a single run within one hour. In pork salami containing horse, turkey, sheep, chicken and beef meat, it was possible to identify species levels as low as 0.01%. In beef salami and cooked pork sausages 0.1% of foreign meat could be detected. This novel workflow enables rapid and sensitive species identification in processed meat at the point of need.

Residual neural networks in single instance-driven identification of fungal pathogens

The rise in fungal infections, attributed to various factors including medical interventions and compromised immune systems, necessitates rapid and accurate identification methods. While traditional mycological diagnostics are time-consuming, machine learning offers a promising alternative. Nevertheless, the scarcity of well-curated datasets is a significant obstacle. To address this, a novel approach for identifying fungi in microscopic images using Residual Neural Networks and a subimage retrieval mechanism is proposed, with the final step involving the implementation of majority voting. The new method, applied to the Digital Images of Fungus Species database, surpassed the original patch-based classification using Convolutional Neural Networks, obtaining an overall classification accuracy of 94.7% compared to 82.4% with AlexNet FV SVM. The observed MCC metric exceeds 0.9, while AUC is near to one. This improvement is attributed to the optimization of hyperparameters and top layer architecture, as well as the effectiveness of the Mish activation function in ResNet-based architectures. Noteworthy, the proposed method achieved 100% accurate classification for images from 8 out of 9 classes after majority voting and is high resistant to overfitting, highlighting its potential for rapid and accurate fungal species identification in medical diagnostics and research.

DNA barcodes and species boundaries of black flies (Diptera: Simuliidae) in Malaysia

Black flies play a prominent role in public health and the epidemiology of parasitic diseases of humans, domesticated and wild animals. Correct identification and a comprehensive survey are required to identify vector and pest species and thus understand their biological attributes which play a vital role in the monitoring program. DNA barcoding is an established molecular tool that provides rapid and accurate species identification. Our study strengthens the molecular database for black flies in Malaysia by adding 59 cytochrome c oxidase I sequences for 22 species, of which 14 are included for the first time. These sequences, combined with those in public databases, represent a total of 338 sequences for 52 Malaysian species, nearly 50% of which were collected from type localities. At the subgeneric level, barcode gap analysis most accurately identified species in the subgenus Nevermannia (92%), followed by Simulium s. l. (91%), and Gomphostilbia (81%). The remaining sequences were ambiguous and could not be distinguished from those of nearest neighbour species due to an overlap in genetic divergence and low genetic diversity, especially between insular species. Tree analyses indicate that certain species had incomplete lineage sorting and low mitochondrial signals. Possible cryptic species were indicated in the Simulium (Gomphostilbia) batoense and S. (G.) epistum species groups. Species delimitations were consistent with morphological identifications except in large species groups such as the S. (G.) asakoae, S. (G.) batoense, S. (G.) epistum, and S. (Simulium) melanopus groups. The use of type specimens or specimens collected from type localities (topotypes) in barcoding is strongly recommended for reference sequences to increase the reliability of the molecular database.

Rapid identification of related species of vibrio by gyrB gene degenerative primers

Vibrio are ubiquitous in the marine environment, and some of them are pathogenic to marine organisms and humans. Due to the complexity and sensitivity of traditional phenotypic and biochemical tests, there is currently no method that could distinguish pathogenic Vibrio on the market. Therefore, this study developed new primers based on gyrB gene to identify Vibrio. We designed 7 pairs of PCR primers (DP1-DP7) and pre-selected them using 7 standard strains of Vibrio. The results showed that the efficiency of DP1 primers (VF1:5′-AARCARGGNCGTAACCGTAA-3′, VR1: 5′-HGGGTADCGRCGRCTCAT-3′) was better. After that, 15 standard strains of Vibrio, 4 non-Vibrio strains and 48 Vibrio strains were used to evaluate the specificity of DP1. Through agarose gel electrophoresis and sequencing analysis of the PCR amplified products. The results showed that the sequence consistency of DP1 primer amplification product was ≥99%, which was consistent with the target strain. This study proved that DP1 primer could be used as a universal primers for Vibrio, providing technical support for rapid screening and identification of Vibrio.

ONTbarcoder 2.0: rapid species discovery and identification with real-time barcoding facilitated by Oxford Nanopore R10.4.

Most arthropod species are undescribed and hidden in specimen-rich samples that are difficult to sort to species using morphological characters. For such samples, sorting to putative species with DNA barcodes is an attractive alternative, but needs cost-effective techniques that are suitable for use in many laboratories around the world. Barcoding using the portable and inexpensive MinION sequencer produced by Oxford Nanopore Technologies (ONT) could be useful for presorting specimen-rich samples with DNA barcodes because it requires little space and is inexpensive. However, similarly important is user-friendly and reliable software for analysis of the ONT data. It is here provided in the form of ONTbarcoder 2.0 that is suitable for all commonly used operating systems and includes a Graphical User Interface (GUI). Compared with an earlier version, ONTbarcoder 2.0 has three key improvements related to the higher read quality obtained with ONT's latest flow cells (R10.4), chemistry (V14 kits) and basecalling model (super-accuracy model). First, the improved read quality of ONT's latest flow cells (R10.4) allows for the use of primers with shorter indices than those previously needed (9 bp vs. 12-13 bp). This decreases the primer cost and can potentially improve PCR success rates. Second, ONTbarcoder now delivers real-time barcoding to complement ONT's real-time sequencing. This means that the first barcodes are obtained within minutes of starting a sequencing run; i.e. flow cell use can be optimized by terminating sequencing runs when most barcodes have already been obtained. The only input needed by ONTbarcoder 2.0 is a demultiplexing sheet and sequencing data (raw or basecalled) generated by either a Mk1B or a Mk1C. Thirdly, we demonstrate that the availability of R10.4 chemistry for the low-cost Flongle flow cell is an attractive option for users who require only 200-250 barcodes at a time.

GTCBS-YOLOv5s: A lightweight model for weed species identification in paddy fields

Accurate and rapid weed species identification contributes to selective herbicide spraying or robotic weeding. An image-based method for automatic identification of weed species in paddy fields is highly desirable. However, water reflection, soil background, occlusion, and varying growth and illumination make the development of this method challenging. To address this issue, an improved deep learning model GTCBS-YOLOv5s was proposed to identify six weed species in paddy fields, where Ghost, C3Trans and convolutional block attention module (CBAM) were employed to improve weed feature extraction in complex environments. The bidirectional feature pyramid network (BiFPN) coupled with Concat structure was introduced in the Neck network to achieve the multi-scale feature fusion for identifying various weed species. Three different output feature maps in Detect network were utilized to identify weeds of varying sizes. A more comprehensive scale-sensitive intersection over union (SIoU) loss function was adopted to eliminate the redundant generating boxes. The results showed that GTCBS-YOLOv5s model achieved a mean average precision (mAP) of 91.1 % for the test set, and the identification speed reached 85.7 FPS. Robustness tests demonstrated that GTCBS-YOLOv5s obtained satisfactory performance in identifying weeds under various lighting conditions, with precision (P), recall (R) and mAP all greater than 85 %. Occluded weeds were identified with P, R and average precision (AP) greater than 89.8 %, 90.1 % and 90.3 %, respectively. Furthermore, GTCBS-YOLOv5s had good performance in identifying weeds at different growth stages, with P, R and mAP higher than 90.1 %, 89.5 %, and 90.3 % respectively. Compared with the state-of-the-art models, GTCBS-YOLOv5s was highly promising for deployment to the embedded devices for real-time field detection due to its high accuracy, lightweight and robust attributes, as well as fast inference.

Effect of image resolution on automatic detection of whitefly (Hemiptera: Aleyrodidae) species on tomato leaflets using deep learning

Deep learning approaches facilitate the rapid and accurate identification of insect pest species in agriculture. However, challenges arise when detecting tiny pests, such as whiteflies (Hemiptera: Aleyrodidae). Although high-resolution images are preferred for accurate detection, they reduce the image acquisition throughput. Hence, this study investigates the impact of image resolution on the accuracy of detecting the adults of two whitefly species, Bemisia tabaci Gennadius and Trialeurodes vaporariorum Westwood, on tomato leaflets. The following state-of-the-art object detectors were used: Faster R-CNN, SSD, YOLOv3, detectoRS, YOLOX, and deformable DETR. The results showed that the models with high-resolution datasets for training and testing performed well in insect detection, although performance declined with decreased image resolution. While, for many models, most errors occurred in species classification, those trained on low-resolution images had a higher proportion of localization errors. Poor performance was also observed when image resolution differed between training and inference. Moreover, the models in which the training dataset was assembled from high- and low-resolution images exhibited similar performance to those trained over each image resolution. Ultimately, detectoRS demonstrated the best performance for lower-resolution images or a mixture of multiple-resolution images, while YOLOX exhibited the highest accuracy for high-resolution images. This study highlights the importance of image resolution for optimal performance in insect pest-detection tasks.