The proteolytic activity in inflammatory bowel disease: insight from gut microbiota.

Inflammatory bowel disease (IBD) is a chronic, recurrent inflammatory disease caused by the destruction of the intestinal mucosal epithelium that affects a growing number of people worldwide. Although the etiology of IBD is complex and still elucidated, the role of dysbiosis and dysregulated proteolysis is well recognized. Various studies observed altered composition and diversity of gut microbiota, as well as increased proteolytic activity (PA) in serum, plasma, colonic mucosa, and fecal supernatant of IBD compared to healthy individuals. The imbalance of intestinal microecology and intestinal protein hydrolysis were gradually considered to be closely related to IBD. Notably, the pivotal role of intestinal microbiota in maintaining proteolytic balance received increasing attention. In summary, we have speculated a mesmerizing story, regarding the hidden role of PA and microbiota-derived PA hidden in IBD. Most importantly, we provided the diagnosis and therapeutic targets for IBD as well as the formulation of new treatment strategies for other digestive diseases and protease-related diseases.

The selective sponging of miRNAs by OIP5-AS1 regulates metabolic reprogramming of pyruvate in adenoma-carcinoma transition of human colorectal cancer.

RNA interactomes and their diversified functionalities have recently benefited from critical methodological advances leading to a paradigm shift from a conventional conception on the regulatory roles of RNA in pathogenesis. However, the dynamic RNA interactomes in adenoma-carcinoma sequence of human CRC remain unexplored. The coexistence of adenoma, cancer, and normal tissues in colorectal cancer (CRC) patients provides an appropriate model to address this issue. Here, we adopted an RNA in situ conformation sequencing technology for mapping RNA-RNA interactions in CRC patients. We observed large-scale paired RNA counts and identified some unique RNA complexes including multiple partners RNAs, single partner RNAs, non-overlapping single partner RNAs. We focused on the antisense RNA OIP5-AS1 and found that OIP5-AS1 could sponge different miRNA to regulate the production of metabolites including pyruvate, alanine and lactic acid. Our findings provide novel perspectives in CRC pathogenesis and suggest metabolic reprogramming of pyruvate for the early diagnosis and treatment of CRC.

Transferability-Guided Cross-Domain Cross-Task Transfer Learning.

We propose two novel transferability metrics fast optimal transport-based conditional entropy (F-OTCE) and joint correspondence OTCE (JC-OTCE) to evaluate how much the source model (task) can benefit the learning of the target task and to learn more generalizable representations for cross-domain cross-task transfer learning. Unlike the original OTCE metric that requires evaluating the empirical transferability on auxiliary tasks, our metrics are auxiliary-free such that they can be computed much more efficiently. Specifically, F-OTCE estimates transferability by first solving an optimal transport (OT) problem between source and target distributions and then uses the optimal coupling to compute the negative conditional entropy (NCE) between the source and target labels. It can also serve as an objective function to enhance downstream transfer learning tasks including model finetuning and domain generalization (DG). Meanwhile, JC-OTCE improves the transferability accuracy of F-OTCE by including label distances in the OT problem, though it incurs additional computation costs. Extensive experiments demonstrate that F-OTCE and JC-OTCE outperform state-of-the-art auxiliary-free metrics by 21.1% and 25.8% , respectively, in correlation coefficient with the ground-truth transfer accuracy. By eliminating the training cost of auxiliary tasks, the two metrics reduce the total computation time of the previous method from 43 min to 9.32 and 10.78 s, respectively, for a pair of tasks. When applied in the model finetuning and DG tasks, F-OTCE shows significant improvements in the transfer accuracy in few-shot classification experiments, with up to 4.41% and 2.34% accuracy gains, respectively.

DOVE: Doodled vessel enhancement for photoacoustic angiography super resolution.

Deep-learning-based super-resolution photoacoustic angiography (PAA) has emerged as a valuable tool for enhancing the resolution of blood vessel images and aiding in disease diagnosis. However, due to the scarcity of training samples, PAA super-resolution models do not generalize well, especially in the challenging in-vivo imaging of organs with deep tissue penetration. Furthermore, prolonged exposure to high laser intensity during the image acquisition process can lead to tissue damage and secondary infections. To address these challenges, we propose an approach doodled vessel enhancement (DOVE) that utilizes hand-drawn doodles to train a PAA super-resolution model. With a training dataset consisting of only 32 real PAA images, we construct a diffusion model that interprets hand-drawn doodles as low-resolution images. DOVE enables us to generate a large number of realistic PAA images, achieving a 49.375% fool rate, even among experts in photoacoustic imaging. Subsequently, we employ these generated images to train a self-similarity-based model for super-resolution. During cross-domain tests, our method, trained solely on generated images, achieves a structural similarity value of 0.8591, surpassing the scores of all other models trained with real high-resolution images. DOVE successfully overcomes the limitation of insufficient training samples and unlocks the clinic application potential of super-resolution-based biomedical imaging.

Piperidine-based ionic liquid additive with electrostatic shielding and redox activity enabling advanced lithium-oxygen batteries.

Here, we propose a piperidine-based ionic liquid additive. The electrostatic shielding effect of the piperidine cation (PP13+) effectively inhibits the growth of lithium dendrites. Simultaneously, the redox activity of the bromine anion synergistically reduces the overpotential. This approach significantly improves the cycling performance of lithium-oxygen batteries.

UCL-Dehaze: Toward Real-World Image Dehazing via Unsupervised Contrastive Learning.

While the wisdom of training an image dehazing model on synthetic hazy data can alleviate the difficulty of collecting real-world hazy/clean image pairs, it brings the well-known domain shift problem. From a different yet new perspective, this paper explores contrastive learning with an adversarial training effort to leverage unpaired real-world hazy and clean images, thus alleviating the domain shift problem and enhancing the network's generalization ability in real-world scenarios. We propose an effective unsupervised contrastive learning paradigm for image dehazing, dubbed UCL-Dehaze. Unpaired real-world clean and hazy images are easily captured, and will serve as the important positive and negative samples respectively when training our UCL-Dehaze network. To train the network more effectively, we formulate a new self-contrastive perceptual loss function, which encourages the restored images to approach the positive samples and keep away from the negative samples in the embedding space. Besides the overall network architecture of UCL-Dehaze, adversarial training is utilized to align the distributions between the positive samples and the dehazed images. Compared with recent image dehazing works, UCL-Dehaze does not require paired data during training and utilizes unpaired positive/negative data to better enhance the dehazing performance. We conduct comprehensive experiments to evaluate our UCL-Dehaze and demonstrate its superiority over the state-of-the-arts, even only 1,800 unpaired real-world images are used to train our network. Source code is publicly available at https://github.com/yz-wang/UCL-Dehaze.

Study on mechanical properties and acoustic emission characteristics of deep diorite under uniaxial compression.

The research object is diorite in the Lingbei TBM section of the Hanjiang-To-Weihe River Qinling tunnel, with a buried depth of over 1 km. Using MTS-2000 microcomputer-controlled electro-hydraulic servo universal testing machine and DS5-16b acoustic emission (AE) monitoring system, uniaxial compression and acoustic emission monitoring tests were carried out on rock samples, to study the uniaxial compression mechanical properties and acoustic emission characteristics of the deep diorite. The results of the study indicate that: (1) During uniaxial compression, diorite undergoes four stages: initial compaction, elasticity, yield and failure, in which the curve of the initial compaction stage is significantly smoother. The uniaxial compressive strength is 41.95 MPã102.42 MPa, with an average of 74.07 Mpa; The axial peak strain ranges from 1 % to 1.4 %, and the failure mode belongs to brittle ductile splitting failure. (2) The cumulative ringing count and energy showed a very slow increase trend during the calm period; After entering a surge period (with the appearance of Kaiser points), both show a significant transition state; During the slow increase period, the overall growth rate of the two slowed down and remained almost silent. (3) On the basis of the analysis of RA-AF values during the deformation and rupture process of diorite, it can be seen that the damage type of diorite is tensile damage by the significant low RA value and high AF value characteristics, which coincides with the actual damage fracture characteristics of the rocks in the sample. (4) During the compaction stage, there are few acoustic emission location points, which correspond to low energy and are mainly distributed at the higher and lower ends of the sample; After entering the elasticity stage, the number of positioning points significantly increases and gradually expands towards the middle; Near Kaiser point, the number of location points and corresponding energy are both in a sharp increase state, and this trend is in good agreement with the changes in the ringing count-time and energy-time curves. (5) The damage time mainly starts at the end of the calm period, and the pattern of change in the damage curve coincides with the localization point and energy evolution. The results of the research can be used as a referential basis for the development of the excavation, protection and other construction plans for the Lingbei TBM section of the Hanjiang-To-Weihe River Qinling tunnel or similar surrounding rock tunnels, as well as for further conducting triaxial unloading tests on diorite.

CAPRL Scoring System for Prediction of 30-day Mortality in 949 Patients with Coronavirus Disease 2019 in Wuhan, China: A Retrospective, Observational Study.

Background: Coronavirus disease 2019 (COVID-19) is a serious and even lethal respiratory illness. The mortality of critically ill patients with COVID-19, especially short term mortality, is considerable. It is crucial and urgent to develop risk models that can predict the mortality risks of patients with COVID-19 at an early stage, which is helpful to guide clinicians in making appropriate decisions and optimizing the allocation of hospital resoureces. Methods: In this retrospective observational study, we enrolled 949 adult patients with laboratory-confirmed COVID-19 admitted to Tongji Hospital in Wuhan between January 28 and February 12, 2020. Demographic, clinical and laboratory data were collected and analyzed. A multivariable Cox proportional hazard regression analysis was performed to calculate hazard ratios and 95% confidence interval for assessing the risk factors for 30-day mortality. Results: The 30-day mortality was 11.8% (112 of 949 patients). Forty-nine point nine percent (474) patients had one or more comorbidities, with hypertension being the most common (359 [37.8%] patients), followed by diabetes (169 [17.8%] patients) and coronary heart disease (89 [9.4%] patients). Age above 50 years, respiratory rate above 30 beats per minute, white blood cell count of more than10 × 109/L, neutrophil count of more than 7 × 109/L, lymphocyte count of less than 0.8 × 109/L, platelet count of less than 100 × 109/L, lactate dehydrogenase of more than 400 U/L and high-sensitivity C-reactive protein of more than 50 mg/L were independent risk factors associated with 30-day mortality in patients with COVID-19. A predictive CAPRL score was proposed integrating independent risk factors. The 30-day mortality were 0% (0 of 156), 1.8% (8 of 434), 12.9% (26 of 201), 43.0% (55 of 128), and 76.7% (23 of 30) for patients with 0, 1, 2, 3, ≥4 points, respectively. Conclusions: We designed an easy-to-use clinically predictive tool for assessing 30-day mortality risk of COVID-19. It can accurately stratify hospitalized patients with COVID-19 into relevant risk categories and could provide guidance to make further clinical decisions.