• 방송공학회논문지
• /
• 제10권4호통권29호
• /
• pp.463-473
• /
• 2005

본 논문에서는 JVT에서 표준화가 진행중인 스케일러블 동영상 부호화의 부호화 효율을 높이기 위해 영상 시퀀스의 시간적변환 특성을 고려하여 적응적으로 GOP의 크기를 선택하여 부호화를 하는 방법을 제안한다. 일반적으로 SVC에서는 한 영상 시퀀스에 대해서 GOP의 크기를 고정하여 부호화를 하는데, GOP의 크기와 영상 시퀀스의 시간적 특성에 따라 부호화 효율이 변하게 된다. 이에 비디오 시퀀스의 지역적인 시간적 특성에 따라 GOP의 크기를 적응적으로 가변 하여 부호화하는 방식을 제안한다. 실험 결과 제안된 적응적 GOP 구조 (Adaptive GOP Structure) 부호화 방법의 부호화 효율이 개선되었으며, Crew 시퀀스에서는 0.63dB 가 향상되었다.

• 방송공학회논문지
• /
• 제18권6호
• /
• pp.816-822
• /
• 2013

최신 동영상 압축 표준화 기술인 HEVC(High Efficiency Video Coding)는 ITU-T(VCEG)와 ISO-IEC(MPEG)에서 JCT-VC(Joint Collaborative Team on Video Coding)라는 팀을 이루어 진행했으며 표준화의 막바지에 다다르고 있다. 기존 H.264/AVC에 약 50% 이상의 성능 향상을 가져왔으나, 다양한 압축 기술을 사용함에 따라 부호화 및 복호화의 복잡도가 매우 증가하는 문제가 있다. 제안하는 방법은 CPU 병렬처리와 GPU 가속처리를 통해 HEVC의 복잡도를 줄이고, 이를 UHD(Ultra High Definition) 초고해상도 영상에 적용하는 방법으로 UHD($3840{\times}2144$) 영상에서 15fps 이상 인코딩/디코딩의 속도를 가지며, CPU와 GPU간의 데이터 전송 방법의 발전으로 추가적인 속도 향상이 기대된다.

• 방송공학회논문지
• /
• 제25권3호
• /
• pp.325-337
• /
• 2020

VVC(Versatile Video Coding)는 ISO/IEC/ITU-T의 JVET(Joint Video Experts Team)에서 표준화 중인 새로운 비디오 부호화 표준으로 스크린 콘텐츠 부호화 툴을 포함한 다양한 기술을 채택하고 있다. 스크린 콘텐츠는 문자 영역과 같이 사선 방향 에지가 자주 발생하는 특징을 가지며, 이런 특징을 갖는 영상에 삼각형 형태의 분할 부호화를 적용하면 압축 효율이 증가할 수 있다. 본 논문에서는 스크린 콘텐츠를 위한 VVC 기반 화면내 삼각형 분할 예측 방법을 제안한다. 기존 VVC의 화면간 예측 부호화에서 삼각형 분할 예측을 지원하는 Triangular Prediction Mode 방법과 유사하게, 제안 방법은 화면내 예측 부호화에서 수직과 수평 방향 예측 모드와 주변 복원 참조 라인을 이용하여 두 개의 사각형 예측 블록을 생성하고 삼각형 모양의 마스크로 두 예측 블록을 가중합하여 최종 예측 신호를 만든다. 제안 방법의 실험 결과는 All Intra 스크린 콘텐츠 영상 실험에서 YUV 각각 평균 1.86%, 1.49%, 1.55% 부호화 성능향상을 보이고, 자연 영상 실험 조건에서는 부호화 효율에 미미한 손실을 보였다. 결론적으로, 화면내 예측 부호화 모드에 제안 방법을 적용하여 압축 성능을 향상할 수 있었다.

• 한국식물병리학회:학술대회논문집
• /
• 한국식물병리학회 2003년도 정기총회 및 추계학술발표회
• /
• pp.76.2-77
• /
• 2003

The complete nucleotide sequences of genomic RNA of an isolate of soybean mosaic virus (SMV-CN18), which has ability to overcome Rsv resistance of soybean, have been determined. A large open reading frame encodes a polyprotein of 3068 amino acids with a predicted Mr of 350 kDa. Based on comparison with the proposed cleavage site of other potyviral polyproteins, nine mature proteins are predicted as a following order, P1, HC-Pro, P3, CI, 6K, VPg, NIa, NIb and coat protein (CP). The mature proteins of the strain share various amino acid identity with known SMV-G2, -G7 and -N strain, with the greatest variability occurring in the P1 (91 ％, 88 ％, 96％)and the lowest variability in the CP (100 ％, 99 ％, 100 ％). In addition, 5' untranslated region determined by 5' RACE is much more various than any coding regions. Difference in amino acid sequences throughout the genome is discussed in relation to resistance and susceptibility of soybean cultivars to SMV-CNl8.

• Journal of Photoscience
• /
• 제12권2호
• /
• pp.75-82
• /
• 2005

Rumex acetosa L. is a dioecious flowering plant with well developed sex chromosome system: 2n = 12 + XX in the female plants and 2n = 12 + XY1Y2 in the male plants. To isolate sex-linked DNA, we carried out chromosome micromanipulation, followed by DOP-PCR, AFLP of the PCR products, reverse Southern hybridization and sequence analysis. From 500 AFLP specific clones, 13 X-chromosome and 5 Y-chromosome specific clones were obtained. Except one clone RADAX-239 ($\underline{R}umex\;\underline{a}-\underline{D}OP-PCR-\underline{A}FLP-\underline{Y}-chromosome\;specific$), all clones appear to be R. acetosa plant-specific sequences and non-coding sequences. Southern blot analysis using these clones could not discriminate genomic DNAs either from male or female plants. Results of this study imply that both autosome-origin and degeneration of sex chromosomes are prevalent in plant systems.

• 한국미생물·생명공학회지
• /
• 제22권2호
• /
• pp.134-142
• /
• 1994

The neucleotide sequences of the xylA gene encoding $\beta $-xylosidase of Bacillus stearothermophilus and is its flanking regions were datermined. Three open reading frame(ORFs) were found, one of which(ORF1) appeared to code for the $\beta $-xylosidase. The 1830 base pair ORF1 encoded 609 amino acids starting from a TTG initiation codon. The molecular weight deduced from the nucleotide sequence(68 KD) was in agreement with that estimated by SDS-polyacrylamide gel electrophoresis of the purified enzyme(66 KD). The Shine-Dalgarno sequence(5'-AGGAGG-3') was found 11 bp upstream of the initiation codon. Further 15 bp upstream, there observed a potential transcription initiation signals. The putative -10 sequence(CATAAT) and -35 sequence(TTGTTA) coresponded closely to the consensus sequences for Bacillus subtilis RNA polymerase with major sigma factor. The guanine-plus-cytosine content of the coding region of the xylA gene was 56mol% while that of the third position of the codons was 63 mol%. Based on the comparison with the amino acid sequences of several other carbohydrate degrading enzymes, two conserved regions, possibly participating in the catalytic mechamism of $\beta $-xylosidase xylA, were identified in 278-298 and 329-350 regions of the translated xylA gene. The nucleotide sequence of the xylA was found to exhibit no homology to any other genes so far reproted.

• Fisheries and Aquatic Sciences
• /
• 제15권4호
• /
• pp.325-335
• /
• 2012

The present study characterized the fast skeletal myosin light chain-2 gene (mlc2f) in the euryhaline Javanese ricefish (Oryzias javanicus: Beloniformes). Coding nucleotide and deduced amino acid sequences of Javanese ricefish mlc2f were well conserved in the vertebrate lineage. Javanese ricefish mlc2f showed a typical seven-exon structure, and its promoter exhibited transcription factor binding motifs common to most muscle-specific genes. However, Javanese ricefish mlc2f also displayed tandem duplications of intronic sequences in both intron 1 and intron 3. Based on quantitative reverse transcription-polymerase chain reaction, the mlc2f transcripts were highly predominant in skeletal muscles of adults and were differentially modulated during embryonic development. Microinjection of the mlc2f promoter-driven red fluorescent protein (RFP) reporter construct successfully exhibited heterologous expression of the fluorescent reporter, primarily in muscular areas of hatchlings, although the distribution pattern of RFP signals was not uniform due to the mosaic nature of the introduced transgene. Data from this study indicate that the Javanese ricefish mlc2f gene has undergone "intra-intronic" duplication events in a species-specific manner and that the mlc2f regulator may also be useful in heterologous expression assays of the skeletal muscles of this species.

• Toxicological Research
• /
• 제8권2호
• /
• pp.331-342
• /
• 1992

The chicken major histocompatibility complex (MHC), the B complex, is beginning to be analyzed at the DNA level. Inbred lines of chickens have been reported to possess 3~5 MHC class II genes. To further analyzed the molecular structure of the chicken MHC class II genes, cDNA clones coding for chicken MHC class II (B-L) ${\beta}$ chain molecules were isolated from chicken spleen and liver. Tissue-specific transcription of B-L ${\beta}$genes was studied by Northern blot analysis. A high level of expression was detected for spleen poly(A)$^+$ RNA whereas a faint signal was detected for liver poly(A)$^+$ RNA. Twenty-nine cDNA clones were isolated from the spleen and eight cDNA clones were isolated from the liver. Based on restriction maps, most clones could be clustered into one family of genes. Four cDNA clones were sequenced (S7, S10 and S19 from the spleen and L1, which was identical to S19, from the liver). Complete amino acid sequences of B-L ${\beta}$ chain molecules were predicated from the nucleotide sequences of the cDNA clones. Although both the nature and the location of the conserved residues were similar in chicken and mammalian sequences, some species-specific differences were found, suggesting that the structures of the B-L molecules are similar, but not identical to their mammalian counterparts.

• Ocean Science Journal
• /
• 제40권3호
• /
• pp.155-166
• /
• 2005

New PCR primers (N=18) were designed for the isolation of complete SSU to LSU rDNA sequences from the dinoflagellate Alexandrium tamarense. Standard PCR, employing each primer set selected for amplifications of less than 1.5 kb, successfully amplified the expected rDNA regions of A. tamarense (Korean isolate, HY970328M). Complete SSU, LSU rDNAs and ITS sequences, including 5.8S rDNA, were recorded at 1,800 bp, 520 bp and 3,393 bp, respectively. The LSU rDNA sequence was the first report in Alexandrium genus. No intron was found in the LSU rRNA coding region. Twelve D-domains within the LSU rDNA were put together into 1,879 bp (44.4% G+C), and cores into 1514 bp (42.8% G+C). The core sequence was significantly different (0.0867 of genetic distance, 91% sequence similarity) in comparison with Prorocentrum micans (GenBank access. no. X16108). The D2 region was the longest in length (300 bp) and highly variable among the 12 D-domains. In a phylogenetic analysis using complete LSU rDNA sequences of a variety of phytoplankton, A. tamarense was clearly separated with high resolution against other species. The result suggests that the sequence may resolve the taxonomic ambiguities of Alexandrium genus, particularly of the tamarensis complex.

• Animal Systematics, Evolution and Diversity
• /
• 제36권4호
• /
• pp.336-346
• /
• 2020

Mitochondrial genome is an important molecule for systematic and evolutionary studies in metazoans. The development of next-generation sequencing (NGS) technique has rapidly increased the number of mitogenome sequences. The process of generating mitochondrial genome based on NGS includes different steps, from DNA preparation, sequencing, assembly, and annotation. Despite the effort to improve sequencing, assembly, and annotation methods of mitogenome, the low quality and/or quantity sequence in the final map can still be generated through the work. Therefore, it is necessary to check and curate mitochondrial genome sequence after annotation for proofreading and feedback. In this study, we introduce the pipeline for sequencing and curation for mitogenome based on NGS. For this purpose, two mitogenome sequences of Dermatobranchus otome were sequenced by Illumina Miseq system with different amount of raw read data. Generated reads were targeted for assembly and annotation with commonly used programs. As abnormal repeat regions present in the mitogenomes after annotation, primers covering these regions were designed and conventional PCR followed by Sanger sequencing were performed to curate the mitogenome sequences. The obtained sequences were used to replace the abnormal region. Following the replacement, each mitochondrial genome was compared with the other as well as the sequences of close species available on the Genbank for confirmation. After curation, two mitogenomes of D. otome showed a typically circular molecule with 14,559 bp in size and contained 13 protein-coding genes, 22 tRNA genes, two rRNA genes. The phylogenetic tree revealed a close relationship between D. otome and Tritonia diomea. The finding of this study indicated the importance of caution and curation for the generation of mitogenome from NGS.