No abstract

Regulating the state of titanium species via the deboronation-assisted route is a facile strategy to construct highly efficient titanosilicate catalysts.

Dinoflagellates are important eukaryotic microorganisms that play critical roles as producers and grazers, and cause harmful algal blooms. The unusual nuclei of dinoflagellates “dinokaryon” have led researchers to investigate their enigmatic nuclear features. Their nuclei are unusual in terms of their permanently condensed nucleosome-less chromatin, immense genome, low protein to DNA ratio, guanine-cytosine rich methylated DNA, and unique mitosis process. Furthermore, dinoflagellates are the only known group of eukaryotes that apparently lack histone proteins. Over the course of evolution, dinoflagellates have recruited other proteins, e.g., histone-like proteins (HLPs), from bacteria and dinoflagellates/viral nucleoproteins (DVNPs) from viruses as histone substitutes. Expression diversity of these nucleoproteins has greatly influenced the chromatin structure and gene expression regulation in dinoflagellates. Histone replacement proteins (HLPs and DVNPs) are hypothesized to perform a few similar roles as histone proteins do in other eukaryotes, i.e., gene expression regulation and repairing DNA. However, their role in bulk packaging of DNA is not significant as low amounts of proteins are associated with the gigantic genome. This review intends to summarize the discoveries encompassing unique nuclear features of dinoflagellates, particularly focusing on histone and histone replacement proteins. In addition, a comprehensive view of the evolution of dinoflagellate nuclei is presented.

No abstract

Background An excessive postoperative inflammatory response is common after surgery for inflammatory bowel disease (IBD) and may be associated with an increased incidence of postoperative ileus. This study assessed the role of perioperative dexamethasone in postoperative ileus after IBD surgery. Method Patients undergoing elective IBD surgery were randomized to either an intravenous 8-mg dose of dexamethasone (n = 151) or placebo (n = 151) upon induction of anesthesia. The primary outcome was the incidence of prolonged postoperative ileus. Secondary outcomes included incidence of reported nausea or vomiting, time to first passage of flatus and stool, GI-2 recovery, postoperative pain, length of stay, and surgical complications. Results An intention-to-treat analysis revealed that patients who received dexamethasone exhibited a lower incidence of prolonged postoperative ileus (22.5% vs 38.4%; P = 0.003), shorter time to first passage of stool (28 vs 48 h, P < 0.001), GI-2 recovery (72 vs 120 h; P < 0.001), reduced postoperative length of stay (9.0 vs 10.0 d; P = 0.002), and less postoperative pain (P < 0.05) compared with controls. Moreover, there were no significant differences in postoperative nausea or vomiting (P = 0.531), major postoperative complications (P = 0.165), or surgical site infections (P = 0.337) between the groups. A benefit was only observed in patients with Crohn’s disease, restored bowel continuity, colon/rectal resections, and those who underwent open operations. Conclusion A single, intravenous 8-mg dose of dexamethasone upon induction of anesthesia reduced the incidence of prolonged postoperative ileus, the intensity of postoperative pain, and shortened the postoperative length of stay for IBD patients undergoing elective surgery. ClinicalTrials.gov: NCT03456752.

Domain generalization (DG) aims to help models trained on a set of source domains generalize better on unseen target domains. The performances of current DG methods largely rely on sufficient labeled data, which however are usually costly or unavailable. While unlabeled data are far more accessible, we seek to explore how unsupervised learning can help deep models generalizes across domains. Specifically, we study a novel generalization problem called unsupervised domain generalization, which aims to learn generalizable models with unlabeled data. Furthermore, we propose a Domain-Irrelevant Unsupervised Learning (DIUL) method to cope with the significant and misleading heterogeneity within unlabeled data and severe distribution shifts between source and target data. Surprisingly we observe that DIUL can not only counterbalance the scarcity of labeled data, but also further strengthen the generalization ability of models when the labeled data are sufficient. As a pretraining approach, DIUL shows superior to ImageNet pretraining protocol even when the available data are unlabeled and of a greatly smaller amount compared to ImageNet. Extensive experiments clearly demonstrate the effectiveness of our method compared with state-of-the-art unsupervised learning counterparts.Recently, the research community has spent significant effort to develop algorithms for learning visual representations without category labels [20,48,61] and shows the advance of contrastive *Corresponing author, also with Beijing Key Lab of Networked Multimedia. Preprint. Under review.

No abstract

No abstract