Motohiro Takagi

種々のアユ種苗が放流されている淀川水系では，海産アユ，琵琶湖産アユのいずれの系統が定着して再生産に貢献しているのか明らかになっていない。淀川水系におけるアユ系統の繁殖貢献および分布を解明するため，産卵場のアユ，河口域および沿岸域のアユ仔稚魚および遡上アユについてマイクロサテライトDNA多型解析による系統判別を行った。いずれの生活期の標本群においても，海産系統に帰属する個体が多かったが，琵琶湖産系統に帰属する個体も認められた。また，琵琶湖産系統のアユが再生産を行っていることが示唆された。

Click on the article title to read more.

A large number of farmed red sea bream (Pagrus major) and their gametes have escaped and dispersed into the wild from net-cage aquaculture since the advent of the technology in the mid-20th century. These escapees have likely been recruited into wild populations, affecting their genetic diversity. In this study, we used assignment methods with microsatellite markers to assess the frequency of escapee farmed red sea bream within the wild populations in a major aquaculture production region (UWJ) and in two regions with relatively lower production (IWG and SIJ) in Ehime Prefecture, Japan. The frequency of escapees in UWJ ranged from 14.1% to 30.2%, as shown by three methods of assignment. Escapees were frequently identified among specimens caught by angling (ranging from 26.7% to 46.7%), suggesting that they stay in the same area even after escape or are less cautious than wild individuals. No escapee was found in IWG and SIJ, areas with less extensive aquaculture. Some wild-caught fish appeared to have hybridized with farmed fish, as indicated by the threshold of membership coefficients obtained by simulation. These results clearly indicate that large numbers of farmed individuals have escaped from UWJ, a major aquaculture area, since the inception of red sea bream farming in Japan. Genetic reshuffle has possibly occurred following hybridization of wild and farmed individuals.

Current image quality assessment (IQA) methods require the original images for evaluation. However, recently, IQA methods that use machine learning have been proposed. These methods learn the relationship between the distorted image and the image quality automatically. In this paper, we propose an IQA method based on deep learning that does not require a reference image. We show that a convolutional neural network with distortion prediction and fixed filters improves the IQA accuracy.

This paper proposes a quality recovery network (QRNet) that recovers the image quality from distorted images and improves the classification accuracy for image classification using these recovered images as the classifier inputs, which are optimized for image quality loss and classification loss. In certain image classification tasks, classifiers based on deep neural networks achieve higher performance compared to those realized by humans. However, these tasks are based on images that are not distorted. To address distorted images, the classifier is fine-tuned with distorted images for practical applications. However, fine-tuning is insufficient for classifying images that include multiple distortion types with severe distortions and often requires the classifier to be retrained for adapting to distorted images, which is a time-consuming process. Therefore, we propose QRNet that generates recovered images for input to the classifier. To address multiple severe distortions, the proposed network is trained using multiple distortion-type images with our proposed loss, which comprises the image quality and classification losses. Moreover, by training the proposed network with multiple classifiers, the recovered images can be easily classified by a new classifier that is not used for training. The new classifier can classify the recovered images without retraining for adapting to distorted images. We evaluate our proposed network with classifiers on public datasets and demonstrate that it improves the classification accuracy for distorted images. Moreover, the experimental results demonstrate that our proposed network with the new classifier improves the classification accuracy.

We previously reported a body color deformity in juvenile red sea bream, which shows transparency in the juvenile stage because of delayed chromatophore development compared with normal individuals, and this finding suggested a genetic cause based on parentage assessments. To conduct marker-assisted selection to eliminate broodstock inheriting the causative gene, developing DNA markers associated with the phenotype was needed. We first conducted SNP mining based on AFLP analysis using bulked-DNA from normal and transparent individuals. One SNP was identified from a transparent-specific AFLP fragment, which significantly associated with transparent individuals. Two alleles (A/G) were observed in this locus, and the genotype G/G was dominantly observed in the transparent groups (97.1%) collected from several production lots produced from different broodstock populations. A few normal individuals inherited the G/G genotype (5.0%), but the A/A and A/G genotypes were dominantly observed in the normal groups. The homologs region of the SNP was searched using a medaka genome database, and intron 12 of the Nell2a gene (located on chromosome 6 of the medaka genome) was highly matched. We also mapped the red sea bream Nell2a gene on the previously developed linkage maps, and this gene was mapped on a male linkage group, LG4-M. The newly found SNP was useful in eliminating broodstock possessing the causative gene of the body color transparency observed in juvenile stage of red sea bream.

アコヤガイ貝殻真珠層の結晶層一層の厚さ（結晶層厚）の遺伝を明らかにするため，2系統の親集団を左殻内面真珠層の色で赤，緑およびそれ以外の3グループに選抜して9家系の次代を生産し，親と子世代における結晶層厚を計測した。また，9家系中2家系および別の4家系の合計6家系における結晶層厚と殻高を継時的に調査した。両親と子世代の平均結晶層厚には強い相関関係が認められたことから，結晶層厚は遺伝形質である可能性が高いと考えられた。一方，各家系の平均結晶層厚は，貝の成長に伴い一定の割合で薄くなることが示唆された。