Jiantao Hu

The Chinese herring (Ilisha elongata) is an economically important clupeiform fish, mostly found along the coast of China and Japan in the western Pacific Ocean. Overfishing and climate change have caused a substantial decline in its annual catch, which may have impacted its population size and structure. Here, we present a chromosome-level genome assembly of the Chinese herring, generated using a combination of Nanopore sequencing, Illumina sequencing, and high-throughput chromatin conformation capture technologies. The total length of this assembly is 802.47 Mbp, anchored to 24 chromosomes, with contig N50 and scaffold N50 values of 13.22 and 33.10 Mbp, respectively. Benchmarking Universal Single-Copy Orthologues analysis indicated high completeness of this assembly, with 94.9% of universal single-copy orthologs. Analysis of resequencing samples from various coastal regions of China and Japan revealed four distinct genetic populations of Chinese herring along the East Asian coast. The Dandong population (Yellow Sea) diverged earliest, exhibiting unique genetic structures and selective sweep signals, whereas the other three populations expanded from a tropical South China Sea ancestral population to the East China Sea and to Japanese waters. Among 23,366 protein-coding genes, several families related to the metabolism of very-long-chain unsaturated fatty acids are particularly contracted compared with the American shad. Additionally, several genes associated with fatty acid synthesis formation have undergone accelerated positive selection. These findings may explain the differential fatty acid ratios in marine versus riverine migratory clupeiforms. This high-quality genome assembly of the Chinese herring may provide valuable resources for comparative study of coastal fishes and benefit the development of a sustainable management strategy for fisheries of the Chinese herring.

This study addresses the challenge of limited access to Her2 testing for breast cancer patients due to high costs. In resource-constrained regions, pathologists heavily rely on cost-effective Hematoxylin and Eosin (H&E) stained images for diagnosis. However, H&E data alone is insufficient for accurate breast cancer typing. Leveraging advancements in deep learning, the paper introduces Unet-like Vision-Transformer GAN (ULViT-GAN), a model integrating Vision Transformer and U-Net with a Harr wavelet module for generating high-pathological-consistency Her2 images from H&E data. Trained on a large dataset, ULViT-GAN outperforms UVCGAN, reducing FID and KID scores by 15.68% and 27.35% (H&E to Her2), and 17.72% and 15.87% (Her2 to H&E). This model offers a cost-effective solution for accurate breast cancer diagnosis, particularly in resource-limited areas, potentially enhancing patient survival rates.