• Proceedings of the National Institute of Ecology of the Republic of Korea
• /
• 제3권2호
• /
• pp.67-72
• /
• 2022

Ecological research relies on the interpretation of large amounts of visual data obtained from extensive wildlife surveys, but such large-scale image interpretation is costly and time-consuming. Using an artificial intelligence (AI) machine learning model, especially convolution neural networks (CNN), it is possible to streamline these manual tasks on image information and to protect wildlife and record and predict behavior. Ecological research using deep-learning-based object recognition technology includes various research purposes such as identifying, detecting, and identifying species of wild animals, and identification of the location of poachers in real-time. These advances in the application of AI technology can enable efficient management of endangered wildlife, animal detection in various environments, and real-time analysis of image information collected by unmanned aerial vehicles. Furthermore, the need for school education and social use on biodiversity and environmental issues using AI is raised. School education and citizen science related to ecological activities using AI technology can enhance environmental awareness, and strengthen more knowledge and problem-solving skills in science and research processes. Under these prospects, in this paper, we compare the results of our early 2013 study, which automatically identified African cichlid fish species using photographic data of them, with the results of reanalysis by CNN deep learning method. By using PyTorch and PyTorch Lightning frameworks, we achieve an accuracy of 82.54% and an F1-score of 0.77 with minimal programming and data preprocessing effort. This is a significant improvement over the previous our machine learning methods, which required heavy feature engineering costs and had 78% accuracy.

• 한국추진공학회:학술대회논문집
• /
• 한국추진공학회 2004년도 제22회 춘계학술대회논문집
• /
• pp.802-806
• /
• 2004

For the protection of the local air quality and the global atmosphere, the emissions of trace species including nitric oxides (NO and NO$_2$) from gas turbines are regulated by local governments and by the International Civil Aviation Organization. In-situ measurements of such species are needed not only for the development of advanced low-emission combustion concepts but also for providing emissions data required for the sound assessment of the effects of the emissions on environment. We have been developing a laser absorption system that has a capability of simultaneous determination of NO and NO$_2$concentrations in the exhaust jets from aero gas turbines. A diode laser operating near 1.8 micrometer is used for the detection of NO while a separated visible tunable diode laser operating near 676 nanometers is used for NO$_2$. The sensitivities at elevated temperature conditions were determined for simulated gas mixtures heated up to 500K in a heated cell of a straight 0.5 m optical path. Sensitivity limits estimated as were 30 ppmv-m and 3.7 ppmv-m for NO and NO$_2$, respectively, at a typical exhaust gas temperature of 800K. Experiments using the simulated exhaust flows have proven that $CO_2$ and $H_2O$ vapor - both major combustion products - do not show any interference in the NO or NO$_2$ measurements. The measurement system has been applied to the NO/NO$_2$ measurements in NO and NO$_2$ doped real combustion gas jets issuing from a rectangular nozzle having 0.4 m optical path. The lower detection limits of the system were considerably decreased by using a multipass optical cell. A pair of off-axis parabola mirrors successfully suppressed the beam steering in the combustion gas jets by centralizing the fluctuating beam in sensor area of the detectors.

• The Plant Pathology Journal
• /
• 제39권2호
• /
• pp.191-206
• /
• 2023

Ground cherry (Physalis pubescens) is the most prominent species in the Solanaceae family due to its nutritional content, and prospective health advantages. It is grown all over the world, but notably in northern China. In 2019 firstly bacterial leaf spot (BLS) disease was identified on P. pubescens in China that caused by both BLS pathogens Xanthomonas euvesicatoria pv. euvesicatoria resulted in substantial monetary losses. Here, we compared whole genome sequences of X. euvesicatoria to other Xanthomonas species that caused BLS diseases for high similarities and dissimilarities in genomic sequences through average nucleotide identity (ANI) and BLAST comparison. Molecular techniques and phylogenetic trees were adopted to detect X. euvesicatoria on P. pubescens using recQ, hrpB1, and hrpB2 genes for efficient and precise identification. For rapid molecular detection of X. euvesicatoria, loop-mediated isothermal amplification, polymerase chain reaction (PCR), and real-time PCR techniques were used. Whole genome comparison results showed that the genome of X. euvesicatoria was more closely relative to X. perforans than X. vesicatoria, and X. gardneri with 98%, 84%, and 86% ANI, respectively. All infected leaves of P. pubescens found positive amplification, and negative controls did not show amplification. The findings of evolutionary history revealed that isolated strains XeC10RQ, XeH9RQ, XeA10RQ, and XeB10RQ that originated from China were closely relative and highly homologous to the X. euvesicatoria. This research provides information to researchers on genomic variation in BLS pathogens, and further molecular evolution and identification of X. euvesicatoria using the unique target recQ gene through advance molecular approaches.

• Mycobiology
• /
• 제51권4호
• /
• pp.195-209
• /
• 2023

The seed borne disease such as bakanae is difficult to control. Crop yield loss caused by bakanae depending on the regions and varieties grown, ranging from 3.0% to 95.4%. Bakanae is an important disease of rice worldwide and the pathogen was identified as Fusarium fujikuroi Nirenberg (teleomorph: Gibberella fujikuroi Sawada). Currently, four Fusaria (F. fujikuroi, F. proliferatum, F. verticillioides and F. andiyazi) belonging to F. fujikuroi species complex are generally known as the pathogens of bakanae. The infection occurs through both seed and soil-borne transmission. When infection occurs during the heading stage, rice seeds become contaminated. Molecular detection of pathogens of bakanae is important because identification based on morphological and biological characters could lead to incorrect species designation and time-consuming. Seed disinfection has been studied for a long time in Korea for the management of the bakanae disease of rice. As seed disinfectants have been studied to control bakanae, resistance studies to chemicals have been also conducted. Presently biological control and resistant varieties are not widely used. The detection of this pathogen is critical for seed certification and for preventing field infections. In South Korea, bakanae is designated as a regulated pathogen. To provide highly qualified rice seeds to farms, Korea Seed & Variety Service (KSVS) has been producing and distributing certified rice seeds for producing healthy rice in fields. Therefore, the objective of the study is to summarize the recent progress in molecular identification, fungicide resistance, and the management strategy of bakanae.

• 한국수산과학회지
• /
• 제36권3호
• /
• pp.239-246
• /
• 2003

We have examined the 16S-23S rRNA intergenic spacer region (ISR) of Vibrio vulnificus KCTC 2959. ISRs were amplified by primers complementary to conserved regions of 16S and 23S rRNA genes. ISR amplicons were cloned and sequenced. Analysis of the ISR sequences showed that V. vulnificus KCTC 2959 contains five types of polymorphic ISRs. Size of ISRs ranged from 424 to 741 bp in length and the number of tRNA genes ranged from one to four. The ISRs were designated as ISR-E $(tRNA^{Glu}),\;ISR-IA\;(tRNA^{Ile}-tRNA^{Ala})$, ISR-EKV $(tRNA^{Glu}-tRNA^{Lys}-tRNA^{Val})$, ISR-IAV $(tRNA^{Ile}-tRNA^{Ala}-tRNA^{val})$ and ISR-EKAV $(tRNA^{Glu}-tRNA^{Lys}-tRNA^{Ala}-tRNA^{Val})$ based on their tRNA genes. Multiple alignment of representative sequences from different Vibrio species revealed several domains of high sequence variability. We used the sequences of variable domains to design species-specific primer for detection PCR. Specificity of the primers was examined using genomic DNA prepared from 18 different Vibrio species. The results showed that the PCR using primers designed in this study can be used to detect V. vulnificus from other Vibrio species.

• The Plant Pathology Journal
• /
• 제35권6호
• /
• pp.654-661
• /
• 2019

The soybean cyst nematode, Heterodera glycines, is a major plant-parasitic nematode that has caused important economic losses to Korea's soybean production. Four species of cyst nematodes, H. schachtii, H. glycines, H. trifolii, and H. sojae, all belong to schachtii group are coexist in field soil in Korea. The rapid identification of the nematode is crucial for preventing crop damage and in decision making for controlling this nematode. This study aimed to develop a species-specific primer set for quantitative PCR (qPCR) assay of H. glycines. The specific primer set (HGF1 and HGR1) for H. glycines was designed based on the cytochrome c oxidase subunit I (COI) sequence of mitochondrial DNA. After optimization, it is possible to identify the H. glycines using a qPCR assay with DNA extracted from a single cyst and single second-stage juvenile (J2). The specificity was confirmed by the absence of SYBR fluorescent signals of three other Heterodera species. A serial dilution of DNA extracted from a single cyst was obtained for the sensitivity test. The result showed that the standard curve of the test had a highly significant linearity between DNA concentration and Ct value (R² = 0.996, slope = -3.49) and that the detection limit concentration of DNA of the primer set was 10 pg of DNA per reaction. Our findings suggested that H. glycines could be distinguished from H. sojae and other Heterodera species when a qPCR assay is used with a specific primer set.

• Journal of Ginseng Research
• /
• 제34권1호
• /
• pp.41-46
• /
• 2010

Expensive herbs such as ginseng are always a possible target for fraudulent labeling. New mountain ginseng strains have occasionally been found deep within mountain areas and commercially traded at exorbitant prices. However, until now, no scientific basis has existed to distinguish such ginseng from commonly cultivated ginseng species other than by virtue of being found within deep mountain areas. Polymerase chain reaction (PCR) analysis of the internal transcribed spacer has been shown to be an appropriate method for the identification of the most popular species (Panax ginseng) in the Panax ginseng genus. A single nucleotide polymorphism (SNP) has been identified between three newly found mountain ginseng (KGD4, KGD5, and KW1) and already established Panax species. Specific PCR primers were designed from this SNP site within the sequence data and used to detect the mountain ginseng strains via multiplex PCR. The established multiplex-PCR method for the simultaneous detection of newly found mountain ginseng strains, Korean ginseng, and foreign ginseng in a single reaction was determined to be effective. This study is the first report of scientific discrimination of "mountain ginsengs" and describes an effective method of identification for fraud prevention and for uncovering the possible presence of other, cheaper ginseng species on the market.

• Journal of Microbiology and Biotechnology
• /
• 제32권8호
• /
• pp.1011-1016
• /
• 2022

Bacillus subtilis is a useful bacterium in the food industry with applications as a starter strain for fermented food and as a probiotic. However, it is difficult to discriminate B. subtilis from other Bacillus species because of high phenotypic and genetic similarity. In this study, we employed five previously constructed multilocus sequence typing (MLST) methods for the discrimination of B. subtilis from other Bacillus species and all five MLST assays clearly distinguished B. subtilis. Additionally, the 17 housekeeping genes used in the five MLST assays also clearly distinguished B. subtilis. The pyruvate carboxylase (pyrA) and shikimate dehydrogenase (aroE) genes were selected for the discrimination of B. subtilis because of their high number of polymorphic sites and the fact that they displayed the lowest homology among the 17 housekeeping genes. Specific primer sets for the pyrA and aroE genes were designed and PCR products were specifically amplified from B. subtilis, demonstrating the high specificity of the two housekeeping genes for B. subtilis. This species-specific PCR method provides a quick, simple, powerful, and reliable alternative to conventional methods in the detection and identification of B. subtilis.

• Advances in nano research
• /
• 제14권4호
• /
• pp.313-327
• /
• 2023

Zinc metal organic framework (MOF_Zn)-loaded goad nanoparticles (AuNPs) sol (Au@MOF_Zn), which was characterized by TEM, Mapping, FTIR, XRD, and molecular spectrum, was prepared conveniently by solvothermal method. The results indicated that Au@MOF_Zn had a very strong catalytic effect with the nanoreaction of AuNPs formation between sodium oxalate (SO) and HAuCl4. AuNPs in the new indicator reaction had a strong resonance Rayleigh scattering (RRS) signal at 370 nm. The indicator AuNPs generated by this reaction, which had the most intense surface enhanced Raman scattering (SERS) peak at 1621 cm -1. The new SERS/RRS indicator reaction in combination with specific aptamer (Apt) to fabricate a sensitive and selective Au@MOF_Zn catalytic amplification-aptamer SERS/RRS assay platform for carbendazim (CBZ), with SERS/RRS linear range of 0.025-0.5 ng/mL. The detection limit was 0.02 ng/mL. Similarly, this assay platform has been also utilized to detect oxytetracycline (OTC) and profenofos (PF).

• The Plant Pathology Journal
• /
• 제20권3호
• /
• pp.200-205
• /
• 2004

The partial nucleotide sequences of the genomic dsRNA mycovirus infecting Pleurotus ostreatus isolates ASI2223 and Suhan were determined and compared with those of mycoviruses belonging to partitiviruses and totiviruses. Partial nucleotide sequences of the purified dsRNA from ASI2223 and Suhan showed RNA-dependent RNA polymerase sequences that are closely related to those of partitiviruses, including Fusarium poae virus 1, Fusarium solani virus, Rhizoctoniasolani virus, Discula destructiva virus 2, and Oyster mushroom isometric virus 2. Specific primers were designed for RT-PCR detection of dsRNA viruses from the P. ostreatus isolate ASI2223 and Suhan. Two virus specific primer sets were found to specifically detect each virus among six sets of designed oligonucleotide primers. Collectively, these results suggest that dsRNA mycoviruses from P. ostreatus isolates ASI2223 and Suhan belong to the family Partitiviridae, although, they are not the same virus species. Our results also suggest that these virus-specific primer sets can be employed for the specific detection of each viral sequence in infected tissues.