Search

We address the source-free domain adaptation (SFDA) problem, where only the source model is available during adaptation to the target domain. We consider two settings: the offline setting where all target data can be visited multiple times (epochs) to arrive at a prediction for each target sample, and the online setting where the target data needs to be directly classified upon arrival. Inspired by diverse classifier based domain adaptation methods, in this paper we introduce a second classifier, but with another classifier head fixed. When adapting to the target domain, the additional classifier initialized from source classifier is expected to find misclassified features. Next, when updating the feature extractor, those features will be pushed towards the right side of the source decision boundary, thus achieving source-free domain adaptation. Experimental results show that the proposed method achieves competitive results for offline SFDA on several benchmark datasets compared with existing DA and SFDA methods, and our method surpasses by a large margin other SFDA methods under online source-free domain adaptation setting. • We propose a new method for source-free domain adaptation. • Our method can also be applied in online setting. • The proposed method obtains good results on several datasets. [ABSTRACT FROM AUTHOR]

Liquid-liquid microextraction (LLME) combined with gas chromatography-mass spectrometry (GC-MS) and headspace-gas chromatography-ion mobility spectrometry (HS-GC-IMS) were used to detect the variations in volatile compounds during the distillation process of strong flavor raw Baijiu. The 3D fingerprint spectrum clearly showed a variation in volatile compounds from different distillation stages, and most alcohol-soluble and low molecular weight compounds showed a downward trend with the extension of distillation time, but water-soluble, high molecular weight and high boiling compounds showed the opposite result. Then, 50 aroma compounds were sniffed and identified by gas chromatography-olfactometry-mass spectrometry (GC-O-MS) and multivariate statistical analysis including principal component analysis (PCA) and partial least squares discriminant analysis (PLS-DA) confirmed four aroma active markers related to classification. Finally, the receiver operating characteristic (ROC) curve was further used to verify that ethyl butanoate and ethyl hexanoate were the most effective difference marker to distinguish the head Baijiu samples and butanoic acid was the most effective difference marker for distinguishing the heart Baijiu samples from the tail Baijiu samples. [Display omitted] • LLME coupled with GC-O-MS was used to analyze the changes of aroma active compounds. • GC-MS and GC-IMS were compared the differences in the detection of compounds. • ROC curve was used to verify the most effective difference markers. [ABSTRACT FROM AUTHOR]

Soy sauce aroma type baijiu (SSAB) is one of the four basic baijiu flavor types. It has an elegant sauce flavor, empty cup aroma, lasting, mellow and sweet entrance, soft with obvious acidity, delicate taste and long aftertaste. However, to date, no study has identified the key aroma compounds in SSAB. This study introduces SSAB and its production process, and summarizes the research status on flavor compounds. Through analysis and review, some assumptions are made to explain why the key aroma compound of SSAB remains unidentified. We provide insights for the analysis of key aroma compounds in SSAB and lay a theoretical foundation for the improvement and rapid development of SSAB quality. • Specific summarizes the research of key volatile compounds in soy sauce aroma type baijiu. • Summarizes the research of off-odor components in soy sauce aroma type baijiu. • First time to analyzes why the key flavor of soy sauce aroma type baijiu have not been discovered. [ABSTRACT FROM AUTHOR]

• The feruloyl esterase activity in different layers of Jiupei at different fermentation times were determined and tracked. • The microorganisms with feruloyl esterase activity were screened and identified in Jiupei of sesame-flavor baijiu. • The conditions of enzyme production were optimized by liquid fermentation. Baijiu is a national beverage in China. Some functional compounds have been identified in baijiu, one of which is ferulic acid. One route for in situ ferulic acid synthesis is catalysed by feruloyl esterase from microorganisms using baijiu raw materials as substrates. However, few studies have focused on microorganisms obtained from baijiu that have feruloyl esterase activity. In this work, microorganisms with feruloyl esterase activity were screened in the fermented raw materials of baijiu. Four different microorganisms with feruloyl esterase activity were identified, and one strain, designated J1, showed high feruloyl esterase activity. The strain J1 belonged to the genus Acetobacter and showed the closest relationship with Acetobacter fabarum and Acetobacter lovaniensis. The feruloyl esterase production capacity of J1 was confirmed, and the fermentation conditions were optimised. The optimal enzyme production conditions for the highest feruloyl esterase activity required an initial fermentation pH of 7.0 at 30 °C, an inoculation volume of 10 %, and culturing for 24 h. [ABSTRACT FROM AUTHOR]

[Display omitted] • Novel, potent and effective EZH2 degraders were developed. • Compound P4 can effectively induce EZH2 protein degradation and inhibit breast cancer cell growth. • Compound P4 can significantly decrease the degree of H3K27me3 in MDA-MB-231 cell line, induce apoptosis and G 0 /G 1 phase arrest in Pfeiffer and MDA-MB-231 cell lines. • Compound P4 significantly inhibited the migratory and invasive capacity of MDA-MB-231 cells at micromolar concentrations in a dose-dependent manner. • Adequate metabolic stability in vitro. EZH2 (enhancer of zeste homolog 2) is one of the most important histone methyltransferases (HMTs), and overexpression of EZH2 can lead to proliferation, migration and angiogenesis of tumor cells. But most of EZH2 inhibitors are only effective against some hematologic malignancies and have poor efficacy against solid tumors. Here, we report the design, synthesis, and evaluation of highly potent proteolysis targeting chimeric (PROTACs) small molecules targeting EZH2. We developed a potent and effective EZH2 degrader P4 , which effectively induced EZH2 protein degradation and inhibited breast cancer cell growth. Further studies showed that P4 can significantly decrease the degree of H3K27me3 in MDA-MB-231 cell line, induce apoptosis and G 0 /G 1 phase arrest in Pfeiffer and MDA-MB-231 cell lines. Therefore, P4 is a potential anticancer molecule for breast cancer treatment. [ABSTRACT FROM AUTHOR]