Search

Background: The hair follicle is a skin accessory organ that regulates hair development, and its activity varies on a regular basis. However, the significance of metabolites in the hair follicle cycle has long been unknown. Results: Targeted metabolomics was used in this investigation to reveal the expression patterns of 1903 metabolites in cashmere goat skin during anagen to telogen. A statistical analysis was used to investigate the potential associations between metabolites and the hair follicle cycle. The findings revealed clear changes in the expression patterns of metabolites at various phases and in various feeding models. The majority of metabolites (primarily amino acids, nucleotides, their metabolites, and lipids) showed downregulated expression from anagen (An) to telogen (Tn), which was associated with gene expression, protein synthesis and transport, and cell structure, which reflected, to some extent, that the cells associated with hair follicle development are active in An and apoptotic in An–Tn. It is worth mentioning that the expression of vitamin D3 and 3,3',5-triiodo-L-thyronine decreased and then increased, which may be related to the shorter and longer duration of outdoor light, which may stimulate the hair follicle to transition from An to catagen (Cn). In the comparison of different hair follicle development stages (An, Cn, and Tn) or feeding modes (grazing and barn feeding), Kyoto Encyclopedia of Genes and Genomes (KEGG) enrichment analysis revealed that common differentially expressed metabolites (DEMs) (2'-deoxyadenosine, L-valine, 2'-deoxyuridine, riboflavin, cytidine, deoxyguanosine, L-tryptophan, and guanosine-5'-monophosphate) were enriched in ABC transporters. This finding suggested that this pathway may be involved in the hair follicle cycle. Among these DEMs, riboflavin is absorbed from food, and the expression of riboflavin and sugars (D-glucose and glycogen) in skin tissue under grazing was greater and lower than that during barn feeding, respectively, suggesting that eating patterns may also alter the hair follicle cycle. Conclusions: The expression patterns of metabolites such as sugars, lipids, amino acids, and nucleotides in skin tissue affect hair follicle growth, in which 2'-deoxyadenosine, L-valine, 2'-deoxyuridine, riboflavin, cytidine, deoxyguanosine, L-tryptophan, and guanosine-5'-monophosphate may regulate the hair follicle cycle by participating in ABC transporters. Feeding practices may regulate hair follicle cycles by influencing the amount of hormones and vitamins expressed in the skin of cashmere goats. [ABSTRACT FROM AUTHOR]

Online Social Networks (OSNs) have blossomed into prevailing transmission channels for images in the modern era. Adversarial examples (AEs) deliberately designed to mislead deep neural networks (DNNs) are found to be fragile against the inevitable lossy operations conducted by OSNs. As a result, the AEs would lose their attack capabilities after being transmitted over OSNs. In this work, we aim to design a new framework for generating robust AEs that can survive the OSN transmission; namely, the AEs before and after the OSN transmission both possess strong attack capabilities. To this end, we first propose a differentiable network termed SImulated OSN (SIO) to simulate the various operations conducted by an OSN. Specifically, the SIO network consists of two modules: (1) a differentiable JPEG layer for approximating the ubiquitous JPEG compression and (2) an encoder-decoder subnetwork for mimicking the remaining operations. Based upon the SIO network, we then formulate an optimization framework to generate robust AEs by enforcing model outputs with and without passing through the SIO to be both misled. Extensive experiments conducted over Facebook, WeChat and QQ demonstrate that our attack methods produce more robust AEs than existing approaches, especially under small distortion constraints; the performance gain in terms of Attack Success Rate (ASR) could be more than 60%. Furthermore, we build a public dataset containing more than 10,000 pairs of AEs processed by Facebook, WeChat or QQ, facilitating future research in the robust AEs generation. The dataset and code are available at https://github.com/csjunjun/RobustOSNAttack.git. [ABSTRACT FROM AUTHOR]