Relation path feature embedding based convolutional neural network method for drug discovery.

BACKGROUND: Drug development is an expensive and time-consuming process. Literature-based discovery has played a critical role in drug development and may be a supplementary method to help scientists speed up the discovery of drugs. METHODS: Here, we propose a relation path features embedding based convolutional neural network model with attention mechanism for drug discovery from literature, which we denote as PACNN. First, we use predications from biomedical abstracts to construct a biomedical knowledge graph, and then apply a path ranking algorithm to extract drug-disease relation path features on the biomedical knowledge graph. After that, we use these drug-disease relation features to train a convolutional neural network model which combined with the attention mechanism. Finally, we employ the trained models to mine drugs for treating diseases. RESULTS: The experiment shows that the proposed model achieved promising results, comparing to several random walk algorithms. CONCLUSIONS: In this paper, we propose a relation path features embedding based convolutional neural network with attention mechanism for discovering potential drugs from literature. Our method could be an auxiliary method for drug discovery, which can speed up the discovery of new drugs for the incurable diseases.

Viability, morphology, and proteome of Mycobacterium smegmatis MSMEG_0319 knockout strain.

Mycobacterium tuberculosis Rv0228, a membrane protein, is predicted as a drug target through computational methods. MSMEG_0319 (MS0319) in Mycobacterium smegmatis mc(2) 155 is the ortholog of Rv0228. To study the effect of MS0319 protein on M. smegmatis, an MS0319 gene knockout strain (ΔMS0319) was generated via a homologous recombination technique in this study. The results showed that the lack of MS0319 protein in mc(2) 155 cells led to the loss of viability at nonpermissive temperature. Scanning electron microscopy and transmission electron microscopy observations showed drastic changes in cellular shape especially cell wall disruption in ΔMS0319 cells. Proteomic analysis of ΔMS0319 cells through LC-MS/MS revealed that 462 proteins had changes in their expressions by lacking MS0319 protein. The M. tuberculosis orthologs of these 462 proteins were found through BLASTp search and functional clustering and metabolic pathway enrichment were performed on the orthologs. The results revealed that most of them were enzymes involved in metabolism of carbohydrates and amino acids, indicating that Rv0228 played an important role in cellular metabolism. All these results suggested Rv0228 as a potential target for development of antituberculosis drugs.

An integrated microfluidic chip for the detection of bacteria - A proof of concept.

We designed a microfluidic chip as a proof of concept for the detection of bacterial DNA. The chip was fabricated with poly-dimethylsiloxane (PDMS). It included a solid phase extraction (SPE) chamber, two separate channels and multiple loop-mediated isothermal amplification (LAMP) chambers. Three bacterial strains (Escherichia coli O157:H7, methicillin-resistant Staphylococcus aureus and methicillin-sensitive S. aureus) were used to test the feasibility of the device. LAMP products were examined directly using a UV light and verified by agarose gel electrophoresis. Using this chip, we successfully detected E. coli O157:H7, MSSA and MRSA in less than 2 h. The detection limit for genes rfbE, spa and mecA (specific to E. coli O157:H7, MSSA and MRSA, respectively) was <10(2) CFU/100 µl. Further work is required to refine this approach and rigorously assess its analytical and diagnostic specificity and sensitivity.

Finite element study on three osteotomy methods for treating thoracolumbar osteoporotic fracture vertebral collapse complicated with neurological dysfunction.

BACKGROUND: Surgical methods for patients with osteoporotic fracture vertebral collapse complicated with neurological dysfunction are still a topic of debate. We designed an improved osteotomy for the treatment of osteoporotic compression fracture patients with neurological dysfunction. Compared with traditional osteotomy methods such as pedicle subtraction osteotomy (PSO) and bone-disc-bone osteotomy (BDBO), the osteotomy range is reduced. Therefore, we use a finite element method to analyze the biomechanical conditions of these three osteotomy methods and provide a mechanical theoretical basis for the surgical treatment of these three osteotomy methods. METHODS: Based on the CT scan of a patient with L1 osteoporotic fracture vertebral collapse and neurological dysfunction, the finite element model was constructed by importing Mimics software, and three different osteotomy models were established. The forces and displacements of internal fixation device, T1-L5 whole segment, T10 vertebral body, and T10/11 intervertebral disc were recorded under different working conditions. RESULTS: The displacement levels of internal fixation device, T1-L5 spine, T10 vertebral body, and T10/11 intervertebral disc in the modified osteotomy group were between BDBO group and PSO group. The stress in BDBO group was concentrated in titanium mesh and its maximum stress was much higher than that in PSO group and modified osteotomy group. The mechanical distribution of T10/11 intervertebral disc showed that the maximum stress distribution of the three osteotomy methods was similar. CONCLUSION: The relatively simple modified osteotomy has certain advantages in stress and displacement. In contrast, the stability of BDBO group was poor, especially in the lumbar intervertebral disc and lumbar body. For this type of osteotomy patients, it is recommended to avoid postoperative flexion so as not to increase the load.

Modified posterior osteotomy for osteoporotic vertebral collapse with neurological dysfunction in thoracolumbar spine: a preliminary study.

OBJECTIVE: The risk of osteoporotic vertebral collapse (OVC) associated with delayed neurological dysfunction (DND) is substantial, and performing surgery for this condition in elderly patients presents challenges. The focus of the current research is on simplifying surgical procedures while maintaining their effectiveness. This study was designed to contribute clinical data supporting the use of modified posterior osteotomy for treating thoracolumbar OVC with DND. The study compares perioperative clinical parameters, imaging data characteristics, and changes in efficacy outcome indicators to provide evidence for the advancement of this technique. METHODS: A total of 12 patients diagnosed with osteoporotic vertebral collapse and neurological dysfunction were included in the study. All patients underwent modified posterior osteotomy. Data regarding perioperative and radiological parameters as well as complications such as surgery duration, blood loss, ASIA grade, VAS, ODI, regional kyphosis angle (RKA), anterior vertebral height ratio (AVHr), and spinal canal clearance ratio (SCCr), were collected retrospectively. These parameters were then analysed to evaluate the clinical efficacy and safety of the modified posterior osteotomy technique. RESULTS: A total of 12 patients were included in the study, with a mean age of 65.5 ± 9.7 years. The average follow-up period was 29.4 ± 5.0 months. The mean operative blood loss was 483.3 ± 142.0 ml, and the average operative time was 3.7 ± 0.7 h. The visual analogue scale (VAS) score decreased from a preoperative value of 5.8 ± 0.7 to a final follow-up value of 1.3 ± 0.8 (P < 0.05), indicating a significant improvement in pain. The ODI decreased from 65.2 ± 6.0 before surgery to 20.5 ± 7.0, indicating a decrease in disability, and the postoperative neurological function showed a significant improvement. Correction of the RKA was observed, with the angle changing from 35.8 ± 10.8° before surgery to 20.0 ± 3.5° after surgery and to 22.5 ± 3.1° at the final follow-up. Similarly, correction of the AVHr was observed, with the height changing from 39.3 ± 18.0 to 63.0 ± 14.3 after surgery and to 53.9 ± 8.9 at the final follow-up. Correction of the SCCr was also observed, with the ratio changing from 54.9 ± 5.4 to 68.1 ± 5.3 after surgery and to 68.68 ± 6.76 at the final follow-up. CONCLUSIONS: Posterior modified osteotomy is an effective treatment for thoracolumbar osteoporotic fractures with OVC combined with DND. It can significantly preserve vertebral height, increase vertebral canal volume, correct kyphotic angle, and improve postoperative neurological function. The simplified osteotomy also offers advantages in terms of operating time, blood loss, postoperative VAS score, and improvement in lumbar function.

The Analgesic Efficacy of Ultrasound-guided Quadratus Lumborum Block Transmuscular or Posterior Approach After Hip Surgery: A Systematic Review and Meta-analysis with Trial Sequential Analysis.

BACKGROUND: No review or meta-analysis exists to elucidate the efficacy and safety of quadratus lumborum block (QLB) on the pain intensity, opioid requirement, and mobilization in patients undergoing hip surgery. This systematic review and meta-analysis of randomized controlled trials were designed to compare QLB with no block or placebo (without other nerve/plexus blocks) for patients undergoing hip surgery. METHODS: Two individual researchers conducted the platform searches on the PubMed, Cochrane Library, and Embase databases from inception to June 12, 2021. Only English publications were included. The pain at rest score at 12 postoperative hours was designated as the primary outcome. Secondary outcomes included rest pain at rest scores at 6 and 24 postoperative hours, dynamic pain score at 6, 12, and 24 postoperative hours, total opioid consumption, postoperative nausea and vomiting, and patient satisfaction. RESULTS: Seven trials including 514 patients were included. When compared with controls, the QLB technique significantly reduced pain at rest scores at 12 hours after surgery (mean difference -1.15, -1.52 to -0.77, P <0.0001). The secondary outcomes were limited by heterogeneity: secondary pain outcomes and opioid consumption were consistently improved with QLB ( P <0.05); patient satisfaction and postoperative nausea and vomiting were similar between the groups based on the Inverse Variance Heterogeneity model ( P >0.05). The overall quality of evidence was moderate. CONCLUSIONS: There is moderate evidence that QLB employment in hip surgery produces significant reduction in pain scores and opioid consumption within 24 hours. QLB appears to be an appropriate option for postoperative analgesia after hip surgery.

Lumped-Parameter Circuit Platform for Simulating Typical Cases of Pulmonary Hypertensions from Point of Hemodynamics.

Pulmonary hypertension (PH) presents unusual hemodynamic states characterized by abnormal high blood pressure in pulmonary artery. The objective of this study is to simulate how the hemodynamics develops in typical PH cases without treatment. A lumped-parameter circuit platform of human circulation system is set up to simulate hemodynamic abnormalities of PH in different etiologies and pathogenesis. Four typical cases are considered, which are distal pulmonary artery stenosis, left ventricular diastolic dysfunction, ventricular septal defect, and mitral stenosis. The authors propose regulation laws for chambers and vessels to adapt the abnormal hemodynamic conditions for each PH case. The occurrence and development of each PH case are simulated over time using the lumped-parameter circuit platform. The blood pressure, blood flow, pressure-volume relations for chambers and vessels are numerically calculated for each case of PH progression. The model results could be a quite helpful to understand the hemodynamic mechanism of typical PHs. Graphical Abstract.

Document-Level Biomedical Relation Extraction Using Graph Convolutional Network and Multihead Attention: Algorithm Development and Validation.

BACKGROUND: Automatically extracting relations between chemicals and diseases plays an important role in biomedical text mining. Chemical-disease relation (CDR) extraction aims at extracting complex semantic relationships between entities in documents, which contain intrasentence and intersentence relations. Most previous methods did not consider dependency syntactic information across the sentences, which are very valuable for the relations extraction task, in particular, for extracting the intersentence relations accurately. OBJECTIVE: In this paper, we propose a novel end-to-end neural network based on the graph convolutional network (GCN) and multihead attention, which makes use of the dependency syntactic information across the sentences to improve CDR extraction task. METHODS: To improve the performance of intersentence relation extraction, we constructed a document-level dependency graph to capture the dependency syntactic information across sentences. GCN is applied to capture the feature representation of the document-level dependency graph. The multihead attention mechanism is employed to learn the relatively important context features from different semantic subspaces. To enhance the input representation, the deep context representation is used in our model instead of traditional word embedding. RESULTS: We evaluate our method on CDR corpus. The experimental results show that our method achieves an F-measure of 63.5%, which is superior to other state-of-the-art methods. In the intrasentence level, our method achieves a precision, recall, and F-measure of 59.1%, 81.5%, and 68.5%, respectively. In the intersentence level, our method achieves a precision, recall, and F-measure of 47.8%, 52.2%, and 49.9%, respectively. CONCLUSIONS: The GCN model can effectively exploit the across sentence dependency information to improve the performance of intersentence CDR extraction. Both the deep context representation and multihead attention are helpful in the CDR extraction task.

Continual exposure to cigarette smoke extracts induces tumor-like transformation of human nontumor bronchial epithelial cells in a microfluidic chip.

INTRODUCTION: Heavy cigarette smoking-related chronic obstructive pulmonary disease is an independent risk factor for lung squamous carcinoma. However, the mechanisms underlying the malignant transformation of bronchial epithelial cells are unclear. METHODS: In our study, human tumor-adjacent bronchial epithelial cells were obtained from 10 cases with smoking-related chronic obstructive pulmonary disease and lung squamous carcinoma and cultured in an established microfluidic chip for continual exposure to cigarette smoke extracts (CSE) to investigate the potential tumor-like transformation and mechanisms. The integrated microfluidic chip included upstream concentration gradient generator and downstream cell culture chambers supplied by flowing medium containing different concentrations of CSE. RESULTS: Our results showed that continual exposure to low doses of CSE promoted cell proliferation whereas to high doses of CSE triggered cell apoptosis. Continual exposure to CSE promoted reactive oxygen species production in human epithelial cells in a dose-dependent manner. More importantly, continual exposure to low dose of CSE promoted the epithelial-to-mesenchymal transition process and anchorage-independent growth, and increased chromosome instability in bronchial epithelial cells, accompanied by activating the GRP78, NF-κB, and PI3K pathways. CONCLUSIONS: The established microfluidic chip is suitable for primary culture of human tumor-adjacent bronchial epithelial cells to investigate the malignant transformation. Continual exposure to low doses of CSE promoted tumor-like transformation of human nontumor bronchial epithelial cells by inducing reactive oxygen species production and activating the relevant signaling.