• 융합정보논문지
• /
• 제10권3호
• /
• pp.76-91
• /
• 2020

최근에 한국 고유종인 미선나무 (Abeliophyllum distichum Nakai; Oleaceae) 품종 옥황1호의 유전체가 성공적으로 해독되었다. Acteoside는 다양한 활성을 가지는 물질이며, 여러개의 생화학합성과정이 제시되어왔고, 이들을 통합 검토하여 정확한 생화학합성과정을 완성하였다. 유전체 데이터로부터 2차대사산물을 예측할 수 있는 MetaPre-AI^TM와 정확한 acteoside 생화학합성과정, InfoBoss Pathway Database를 활용하여, acteoside에 관여하는 모든 효소의 유전자를 옥황1호 유전체로부터 성공적으로 확인하였다. 이는 옥황1호는 acteoside 물질을 생산할 수 있는 가능성이 있음을 의미한다. 이에 고성능액체크로마토그래피를 사용하여 옥황1호의 캘러스 세포를 분석하여 acteoside과 이의 유도체인 isoacteoside를 확인하였다. 본 연구는 MetaPre-AI^TM은 유전체로부터 2차대사산물을 성공적으로 예측하였다. 이 방법은 화학물질보다 안정적인 DNA를 분석하여 2차 대사산물을 예측하는 효율적인 방법이 될 것이다.

• Journal of information and communication convergence engineering
• /
• 제22권3호
• /
• pp.221-230
• /
• 2024

The aim of this study is to investigate the automatic recognition and analysis of Jeju marine-life images using artificial intelligence (AI) technology. The dataset of marine-life images was prepared using tools such as Python, TensorFlow, and Google Colab (Google Colaboratory). We also developed models by training deep learning AI in image recognition to automatically recognize the species found in these images and extract their associated information, such as taxonomy, characteristics, and distribution. This study is innovative in that it uses deep learning technology combined with imagerecognition technology for marine biodiversity research. In addition, these results will lead to the development of the marine-life industry in Jeju by supporting marine environment monitoring and marine resource conservation. Furthermore, this study is anticipated to contribute to academic advancement, specifically in the study of marine species diversity.

• 한국산업융합학회 논문집
• /
• 제22권6호
• /
• pp.749-756
• /
• 2019

Big data is structured and unstructured data that is so difficult to collect, store, and so on due to the huge amount of data. Many institutions, including universities, are building student convergence systems to foster talents for data science and AI convergence, but there is an absolute lack of research on what kind of education is needed and what kind of education is required for students. Therefore, in this paper, after conducting the correlation analysis based on the questionnaire on basic surveys and courses to improve the curriculum by grasping the satisfaction and demands of the participants in the "2019 Big Data Youth Talent Training Course" held at K University, Regression analysis was performed. As a result of the study, the higher the satisfaction level, the satisfaction with class or job connection, and the self-development, the more positive the evaluation of program efficiency.

• 한국전자통신학회논문지
• /
• 제18권1호
• /
• pp.107-114
• /
• 2023

본 연구는 건축 부재 사용량 예측을 위한 인공지능 기반의 학습모델을 설계 및 구현하는 방법에 대하여 기술하였다. 인공지능(Artifical intelligence : AI) 은 기술의 발전에 힘입어 다양한 분야에서 폭넓게 활용되고 있지만, 건축설계분야 데이터의 특수성 및 빅데이터 수집의 어려움으로 인해 현장 활용도가 매우 저조한 상태이다. 따라서 건축설계분야에서 인공지능 기술을 도입할 수 있도록 건축 부재 단위의 AI문제를 발굴해 내었으며, 해당분야 데이터가 가지는 특이성을 해결하기 위한 새로운 전처리 기법을 고안하였다. 고안된 전처리 기법을 토대로 인공지능 모델을 구현하였고, 구현된 인공지능 모델의 건축 부재 사용량 예측 정확도가 실제 산업에 사용할 수 있는 수준임을 확인하였다.

• 산업과 과학
• /
• 제3권2호
• /
• pp.17-22
• /
• 2024

최근 ICT분야가 다양한 환경에서 사용되면서 지속가능한 농업 환경에서는 ICT 기술들을 활용하여 농작물별 병충해 분석, 농작물 수확시 로봇 사용, 빅 데이터로 인한 예측 등이 가능해졌다. 그러나, 지속 가능한 농업 환경에서는 자원의 고갈, 농업 인구 감소, 빈곤 증가, 환경 파괴 등을 해결하기 위한 노력이 꾸준히 요구되고 있다. 본 연구에서는 지속 가능한 농업 환경 기반의 농작물의 생산 비용 감소 및 효율성을 증가하기 위한 인공지능 기반 빅 데이터 처리 기법을 제안한다. 제안 기법은 AI를 결합한 농작물의 빅 데이터를 처리함으로써 데이터의 보안성과 신뢰성을 강화하고, 더 나은 의사 결정과 비즈니스 가치 추출이 가능하다. 이는 다양한 산업과 분야에서 혁신적인 변화를 이끌어내고, 데이터 중심의 비즈니스 모델의 발전을 촉진할 수 있다. 실험과정에서 제안 기법은 다량의 데이터가 생성되나, 일일이 정답을 태깅하기 힘든 농장 현장에서, 소량의 데이터에 대해서만 정확한 정답을 부여하고, 정답이 부여되지 않은 다량의 데이터와 함께 학습하여, 다량의 정답 데이터로 학습했을 때와 유사한 성능(오차율:0.05 이내)이 나타났다.

• 농업과학연구
• /
• 제49권3호
• /
• pp.407-420
• /
• 2022

Chicken dominates meat consumption because it is low in fat and high in protein and has less or no religious and cultural barriers. Recently, meat quality traits have become the focus of the poultry industry more than ever. Currently, poultry farming is focusing on meat quality to satisfy meat consumer preferences, which are mostly based on high-quality proteins and a low proportion of saturated fatty acids. Meat quality traits are polygenic traits controlled by many genes. Thus, it is difficult to improve these traits using the conventional selection method because of their low to moderate heritability. These traits include pH, colour, drop loss, tenderness, intramuscular fat (IMF), water-holding capacity, flavour, and many others. Genome-wide association studies (GWAS) are an efficient genomic tool that identifies the genomic regions and potential candidate genes related to meat quality traits. Due to their impact on the economy, meat quality traits are used as selection criteria in breeding programs. Various genes and markers related to meat quality traits in chickens have been identified. In chickens, GWAS have been successfully done for intramuscular fat (IMF) content, ultimate pH (pHu) and meat and skin colour. Moreover, GWAS have identified 7, 4, 4 and 6 potential candidate genes for IMF, pHu, meat colour and skin colour, respectively. Therefore, the current review summarizes the significant genes identified by genome-wide association studies for meat quality traits in chickens.

• Animal Bioscience
• /
• 제37권6호
• /
• pp.993-1000
• /
• 2024

Objective: This study was conducted to investigate the differential expression of the major histocompatibility complex (MHC) gene region in Eimeria-infected broiler. Methods: We profiled gene expression of Eimeria-infected and uninfected ceca of broilers sampled at 4, 7, and 21 days post-infection (dpi) using RNA sequencing. Differentially expressed genes (DEGs) between two sample groups were identified at each time point. DEGs located on chicken chromosome 16 were used for further analysis. Kyoto encyclopedia of genes and genomes (KEGG) pathway analysis was conducted for the functional annotation of DEGs. Results: Fourteen significant (false discovery rate <0.1) DEGs were identified at 4 and 7 dpi and categorized into three groups: MHC-Y class I genes, MHC-B region genes, and non-MHC genes. In Eimeria-infected broilers, MHC-Y class I genes were upregulated at 4 dpi but downregulated at 7 dpi. This result implies that MHC-Y class I genes initially activated an immune response, which was then suppressed by Eimeria. Of the MHC-B region genes, the DMB1 gene was upregulated, and TAP-related genes significantly implemented antigen processing for MHC class I at 4 dpi, which was supported by KEGG pathway analysis. Conclusion: This study is the first to investigate MHC gene responses to coccidia infection in chickens using RNA sequencing. MHC-B and MHC-Y genes showed their immune responses in reaction to Eimeria infection. These findings are valuable for understanding chicken MHC gene function.

• Journal of Animal Science and Technology
• /
• 제65권1호
• /
• pp.57-68
• /
• 2023

Flavor is an important sensory trait of chicken meat. The free amino acid (FAA) and nucleotide (NT) components of meat are major factors affecting meat flavor during the cooking process. As a genetic approach to improve meat flavor, we performed a genome-wide association study (GWAS) to identify the potential candidate genes related to the FAA and NT components of chicken breast meat. Measurements of FAA and NT components were recorded at the age of 10 weeks from 764 and 767 birds, respectively, using a White leghorn and Yeonsan ogye crossbred F2 chicken population. For genotyping, we used 60K Illumina single-nucleotide polymorphism (SNP) chips. We found a total of nine significant SNPs for five FAA traits (arginine, glycine, lysine, threonine content, and the essential FAAs and one NT trait (inosine content), and six significant genomic regions were identified, including three regions shared among the essential FAAs, arginine, and inosine content traits. A list of potential candidate genes in significant genomic regions was detected, including the KCNRG, KCNIP4, HOXA3, THSD7B, and MMUT genes. The essential FAAs had significant gene regions the same as arginine. The genes related to arginine content were involved in nitric oxide metabolism, while the inosine content was possibly affected by insulin activity. Moreover, the threonine content could be related to methylmalonyl-CoA mutase. The genes and SNPs identified in this study might be useful markers in chicken selection and breeding for chicken meat flavor.

• Journal of Animal Science and Technology
• /
• 제65권5호
• /
• pp.912-921
• /
• 2023

Genetic diversity analysis is crucial for maintaining and managing genetic resources. Several studies have examined the genetic diversity of Korean domestic chicken (KDC) populations using microsatellite markers, but it is difficult to capture the characteristics of the whole genome in this manner. Hence, this study analyzed the genetic diversity of several KDC populations using high-density single nucleotide polymorphism (SNP) genotype data. We examined 935 birds from 21 KDC populations, including indigenous and adapted Korean native chicken (KNC), Hyunin and Jeju KDC, and Hanhyup commercial KDC populations. A total of 212,420 SNPs of 21 KDC populations were used for calculating genetic distances and fixation index, and for ADMIXTURE analysis. As a result of the analysis, the indigenous KNC groups were genetically closer and more fixed than the other groups. Furthermore, Hyunin and Jeju KDC were similar to the indigenous KNC. In comparison, adapted KNC and Hanhyup KDC populations derived from the same original species were genetically close to each other, but had different genetic structures from the others. In conclusion, this study suggests that continuous evaluation and management are required to prevent a loss of genetic diversity in each group. Basic genetic information is provided that can be used to improve breeds quickly by utilizing the various characteristics of native chickens.

• 한국멀티미디어학회논문지
• /
• 제24권2호
• /
• pp.178-185
• /
• 2021

Three-dimensional high-resolution magnetic resonance imaging (MRI) provides fine-level anatomical information for disease diagnosis. However, there is a limitation in obtaining high resolution due to the long scan time for wide spatial coverage. Therefore, in order to obtain a clear high-resolution(HR) image in a wide spatial coverage, a super-resolution technology that converts a low-resolution(LR) MRI image into a high-resolution is required. In this paper, we propose a super-resolution technique through filter learning based on information on the surrounding gradient information in 3D space from 3D MRI images. In the learning step, the gradient features of each voxel are computed through eigen-decomposition from 3D patch. Based on these features, we get the learned filters that minimize the difference of intensity between pairs of LR and HR images for similar features. In test step, the gradient feature of the patch is obtained for each voxel, and the filter is applied by selecting a filter corresponding to the feature closest to it. As a result of learning 100 T1 brain MRI images of HCP which is publicly opened, we showed that the performance improved by up to about 11% compared to the traditional interpolation method.