Cross-dependent graph neural networks for molecular property prediction.

MOTIVATION: The crux of molecular property prediction is to generate meaningful representations of the molecules. One promising route is to exploit the molecular graph structure through graph neural networks (GNNs). Both atoms and bonds significantly affect the chemical properties of a molecule, so an expressive model ought to exploit both node (atom) and edge (bond) information simultaneously. Inspired by this observation, we explore the multi-view modeling with GNN (MVGNN) to form a novel paralleled framework, which considers both atoms and bonds equally important when learning molecular representations. In specific, one view is atom-central and the other view is bond-central, then the two views are circulated via specifically designed components to enable more accurate predictions. To further enhance the expressive power of MVGNN, we propose a cross-dependent message-passing scheme to enhance information communication of different views. The overall framework is termed as CD-MVGNN. RESULTS: We theoretically justify the expressiveness of the proposed model in terms of distinguishing non-isomorphism graphs. Extensive experiments demonstrate that CD-MVGNN achieves remarkably superior performance over the state-of-the-art models on various challenging benchmarks. Meanwhile, visualization results of the node importance are consistent with prior knowledge, which confirms the interpretability power of CD-MVGNN. AVAILABILITY AND IMPLEMENTATION: The code and data underlying this work are available in GitHub at https://github.com/uta-smile/CD-MVGNN. SUPPLEMENTARY INFORMATION: Supplementary data are available at Bioinformatics online.

Enhancement of surface graft density of MPEG on alginate/chitosan hydrogel microcapsules for protein repellency.

Alginate/chitosan/alginate (ACA) hydrogel microcapsules were modified with methoxy poly(ethylene glycol) (MPEG) to improve protein repellency and biocompatibility. Increased MPEG surface graft density (n(S)) on hydrogel microcapsules was achieved by controlling the grafting parameters including the buffer layer substrate, membrane thickness, and grafting method. X-ray photoelectron spectroscopy (XPS) model was employed to quantitatively analyze n(S) on this three-dimensional (3D) hydrogel network structure. Our results indicated that neutralizing with alginate, increasing membrane thickness, and in situ covalent grafting could increase n(S) effectively. ACAC(PEG) was more promising than ACC(PEG) in protein repellency because alginate supplied more -COO(-) negative binding sites and prevented MPEG from diffusing. The n(S) increased with membrane thickness, showing better protein repellency. Moreover, the in situ covalent grafting provided an effective way to enhance n(S), and 1.00 ± 0.03 chains/nm(2) was achieved, exhibiting almost complete immunity to protein adsorption. This antifouling hydrogel biomaterial is expected to be useful in transplantation in vivo.

Spray-spinning: a novel method for making alginate/chitosan fibrous scaffold.

The subject of our investigations was the process of obtaining alginate/chitosan polyelectrolyte complex (PEC) fibers. In this study, a novel method named "spray-spinning" was developed for the making of these hybrid fibers. In spray-spinning, a chitosan solution was sprayed into a flowing sodium alginate solution and sheared into streamlines. The elongated streamlines subsequently transformed into alginate/chitosan PEC fibers. Average diameter of the fibers increased with the increasing of chitosan concentration used in spinning. The fibers showed a high water-absorbability of about 45 folds of water to their dry weight and retained their integrity after incubation in Minimum Essential Medium (MEM) for up to 30 days. In vitro co-culture experiments indicated that the fibers could support the three-dimensional growth of HepG2 cells and did not display any cyto-toxicity. Moreover, in vivo implanting experiments indicated that the connective tissue cells infiltrated into the implanted fibrous scaffolds in 3 weeks after surgery. These results demonstrated the potential applications of the as-spun fibers in regenerative medicine and tissue engineering.

[Self-microemulsifying drug delivery system increasing solubility and intestinal absorption in situ of tanshinones].

OBJECTIVE: Study the effect of self-microemulsifying drug delivery system(SMEDDS) on the solubility and absorption of tanshinones to guide the selection of composition of tanshinone SMEDDS. METHOD: The solubility of tanshinones in the solution of SMEDDS was determined by UV-spectrometer and the absorption of tanshinone SMEDDS was determined by HPLC as the detection method. RESULT: The solubility of tanshinones in solution of SMEDDS was 10 times in water and 2.5 times in micelle solution. The solubility of tanshinones in solution of SMEDDS was increased with the increasing of oil (MCT) in composition of tanshinone SMEDDS. The absorption constants (Ka) in SMEDDS and micelle solution was 0.479 h(-1) and 0.326 h(-1) respectively, and the absorption half life (t1/2) was 1.44 h and 2.12 h respectively. The absorption was increased with the oil increasing in composition of tanshinone SMEDDS. CONCLUSION: SMEDDS can increase the solubility and absorption of tanshinones significantly and the increasing of oil content (MCT) in SMEDDS composition promote the dissolution and absorption of tanshinones.

The effect of pulmonary vessel suppression on computerized detection of nodules in chest CT scans.

PURPOSE: In chest computed tomography (CT) scans, pulmonary vessel suppression can make pulmonary nodules more evident, and therefore may increase the detectability of early lung cancer. The purpose of this study was to develop a computer-aided detection (CAD) system with a vessel suppression function and to verify the effectiveness of the vessel suppression on the performance of the pulmonary nodule CAD system. METHODS: A CAD system with a vessel suppression function capable of suppressing vessels and detecting nodules was developed. First, a convolutional neural network (CNN)-based pulmonary vessel suppression technique was employed to remove the vessels from lungs while preserving the nodules. Then, a CNN-based pulmonary nodule detector was utilized to sequentially generate nodule candidates and reduce false positives (FPs). The performance levels of CAD systems with and without the vessel suppression function were compared using 888 three-dimensional chest CT scans from the Lung Nodule Analysis 2016 (LUNA16) dataset. The pulmonary nodule detection results were quantitatively evaluated using the average sensitivity at seven predefined FP rates: 0.125, 0.25, 0.5, 1, 2, 4, and 8 FPs per scan. RESULTS: The developed pulmonary nodule CAD system improved the average sensitivity to 0.977 from 0.950 owing to the addition of the vessel suppression function. CONCLUSIONS: The vessel suppression function considerably improved the performance of the CAD system for pulmonary nodule detection. In practice, it would be embedded in CAD systems to assist radiologists in detecting pulmonary nodules in chest CT scans.

[FTIR analysis of the enzymatic degradation of chitosan blend membranes].

Chitosan blend membranes were prepared by casting the mixture solution of highly deacetylated chitosan (HDC) and moderately deacetylated chitosan (MDC). FTIR was used to investigate the components of the blend membranes before and after lysozymic degradation. It was found that the ratio of MDC in the blen d membranes had a linear relation with the degree of deacetylation (DD) of the membranes. In enzymatic degradation process, DD of the membranes exhibited an increasing tendency. The FTIR data indicate that MDC component in the blend membranes can be removed by selective enzymatic degradation. This also suggests that FTIR can be used as an efficient and rapid method to investigate the degradation process of chitosan.

Hydro-spinning: a novel technology for making alginate/chitosan fibrous scaffold.

Alginate/chitosan polyelectrolyte complex (PEC) hybrid fibers are promising materials for scaffold-making in tissue engineering. In this study, a new method termed "hydro-spinning" was developed to make alginate/chitosan hybrid fibers. In hydro-spinning, a chitosan solution was pumped into a flowing sodium alginate solution and sheared into streamlines. These elongated streamlines subsequently transformed into alginate/chitosan PEC ribbon-like fibers before breaking up into pieces. Average diameter and chitosan content of the fibers correlated positively with the chitosan concentration used in spinning. These hybrid fibers showed a high water-absorbability of around 50-fold to 60-fold of water to their dry weight and could retain their integrity after saturation in minimum essential medium (MEM) medium for 30 days. In vitro culture experiments demonstrated that these fibers were able to support the three-dimensional growth of MCF-7, suggesting the potential applications of these fibers in biomedical and bioengineering fields such as tissue engineering.

Effect of surface wettability and charge on protein adsorption onto implantable alginate-chitosan-alginate microcapsule surfaces.

Alginate-chitosan-alginate (ACA) microcapsules have been developed as a device for the transplantation of living cells. However, protein adsorption onto the surface of microcapsules immediately upon their implantation decides their ultimate biocompatibility. In this work, the chemical composition of the ACA membranes was determined using X-ray photoelectron spectroscopy (XPS). The surface wettability and charge were determined by contact angle and zeta potential measurements, respectively. Then, the effects of surface wettability and charge on bovine fibrinogen (Fgn) and gamma globulin (IgG) adsorption onto ACA microcapsules were evaluated. The results showed that ACA microcapsules had a hydrophilic membrane. So, the surface wettability of ACA microcapsules had little effect on protein adsorption. There was a negative zeta potential of ACA microcapsules which varies with the viscosity or G content of alginate used, indicating a varying degree of net negatively charged groups on the surface of ACA microcapsules. The amount of adsorbed protein increased with increasing of positive charge. Furthermore, the interaction between proteins and ACA microcapsules is dominated by electrostatic repulsion at pH 7.4 and that is of electrostatic attraction at pH 6.0. This work could help to explain the bioincompatibility of ACA microcapsules and will play an important role in the optimization of the microcapsule design.

[Study on homogeneity of enzymatic degradation of chitosan as biomaterials by gel permeation chromatography].

Chitosan is an important biomedical material, and its degree of deacetylation is a main parameter of its biodegradation. Gel permeation chromatography was used to investigate the lysozymic degradation of two types of chitosan samples (A and B) with similar degree of deacetylation and relative molecular mass but with different distributions of two units of N-acetyl-D-glucosamine and D-glucosamine. Weight average relative molecular mass, polydispersity and gel permeation chromatograms during the degradation process were obtained. It was found that chitosan sample A with random distribution of the two units underwent a homogeneous degradation process while chitosan sample B with block distribution underwent a heterogeneous degradation process. The results suggest that the homogeneity of the degradation of chitosan materials by lysozyme depends on the distribution type of the two units, which can help to design chitosan-based biomedical devices.