FGTN: Fragment-based graph transformer network for predicting reproductive toxicity.

Reproductive toxicity is one of the important issues in chemical safety. Traditional laboratory testing methods are costly and time-consuming with raised ethical issues. Only a few in silico models have been reported to predict human reproductive toxicity, but none of them make full use of the topological information of compounds. In addition, most existing atom-based graph neural network methods focus on attributing model predictions to individual nodes or edges rather than chemically meaningful fragments or substructures. In current studies, we develop a novel fragment-based graph transformer network (FGTN) approach to generate the QSAR model of human reproductive toxicity by considering internal topological structure information of compounds. In the FGTN model, the compound is represented by a graph architecture using fragments to be nodes and bonds linking two fragments to be edges. A super molecule-level node is further proposed to connect all fragment nodes by undirected edges, obtaining global molecular features from fragment embeddings. The FGTN model achieved an accuracy (ACC) of 0.861 and an area under the receiver operating characteristic curve (AUC) value of 0.914 on nonredundant blind tests, outperforming traditional fingerprint-based machine learning models and atom-based GCN model. The FGTN model can attribute toxic predictions to fragments, generating specific structural alerts for the positive compound. Moreover, FGTN may also have the capability to distinguish various chemical isomers. We believe that FGTN can be used as a reliable and effective tool for human reproductive toxicity prediction in contribution to the advancement of chemical safety assessment.

Comparison of the outcomes of monopolar and bipolar radiofrequency ablation in surgical treatment of atrial fibrillation.

OBJECTIVE: To compare the therapeutic effects and safety of monopolar and bipolar radiofrequency (RF) ablation used during cardiac surgery to treat atrial fibrillation. METHODS: We retrospectively studied a total of 81 patients with chronic atrial fibrillation who underwent open cardiac surgery with concomitant RF ablation between January 2007 and March 2011. Fifty-eight patients received bipolar RF ablation and 23 received monopolar RF ablation, respectively. The sinus rhythm restoration rate, the procedural duration, the frequency of severe perioperative complications, and mortality were compared between the two groups. RESULTS: The sinus rhythm restoration rate did not differ significantly between the two groups after follow-up of 15.1 ± 12.6 months (P=0.199). The frequencies of severe perioperative complications and mortality were also similar in the two groups. The total procedural time using bipolar RF ablation was significantly shorter than that using monopolar ablation (19.7 ± 4.6 minutes vs. 28.1 ± 8.5 minutes, P< 0.001). CONCLUSIONS: Both monopolar and bipolar RF ablation are safe and effective in treating chronic atrial fibrillation patients during open cardiac surgery, but bipolar RF ablation is more convenient in practice.

[Impact of road infrastructure on ecological networks in the Guangdong-Hong Kong-Macao Greater Bay Area, China]. / 粤港澳大湾区道路基础设施对生态网络的影响.

The construction of road infrastructure has resulted in the degradation of wildlife habitat and the decrease of ecological network connectivity and stability. Studying the impacts of road infrastructure on wildlife life and migration is significant for regional wildlife conservation and ecological network optimization. We assessed the impacts of road infrastructure on habitat suitability using the MaxEnt model based on wildlife occurrence point data in the Guangdong-Hong Kong-Macao Greater Bay Area. We constructed the ecological networks and identified ecological breakpoints using the minimum cumulative resistance model, and compared the ecological network connectivity of different scenarios with the landscape connectivity index and graph theory index. The results showed that railway and motorway significantly affected habitat suitability, causing a decrease in wildlife habitat suitability. Affected by road infrastructure, the fragmentation of ecological sources intensified, the resistance of ecological corridors increased, and the ecological network connectivity and stability significantly decreased. A total of 536 ecological breakpoints were identified, which were concentrated in the area adjacent to ecological sources. The results would provide important scientific references for wildlife habitat conservation and ecological restoration in the Guangdong-Hong Kong-Macao Greater Bay Area.

Comprehensive hepatotoxicity prediction: ensemble model integrating machine learning and deep learning.

Background: Chemicals may lead to acute liver injuries, posing a serious threat to human health. Achieving the precise safety profile of a compound is challenging due to the complex and expensive testing procedures. In silico approaches will aid in identifying the potential risk of drug candidates in the initial stage of drug development and thus mitigating the developmental cost. Methods: In current studies, QSAR models were developed for hepatotoxicity predictions using the ensemble strategy to integrate machine learning (ML) and deep learning (DL) algorithms using various molecular features. A large dataset of 2588 chemicals and drugs was randomly divided into training (80%) and test (20%) sets, followed by the training of individual base models using diverse machine learning or deep learning based on three different kinds of descriptors and fingerprints. Feature selection approaches were employed to proceed with model optimizations based on the model performance. Hybrid ensemble approaches were further utilized to determine the method with the best performance. Results: The voting ensemble classifier emerged as the optimal model, achieving an excellent prediction accuracy of 80.26%, AUC of 82.84%, and recall of over 93% followed by bagging and stacking ensemble classifiers method. The model was further verified by an external test set, internal 10-fold cross-validation, and rigorous benchmark training, exhibiting much better reliability than the published models. Conclusion: The proposed ensemble model offers a dependable assessment with a good performance for the prediction regarding the risk of chemicals and drugs to induce liver damage.

Adjuvant immunotherapy in early-stage resectable non-small cell lung cancer: A new milestone.

Currently, chemotherapy is the standard adjuvant treatment for early-stage non-small cell lung cancer (NSCLC). However, adjuvant cisplatin-based chemotherapy after surgery has been shown to improve 5-year survival rates by only 4-5%. Immunotherapy using immune checkpoint inhibitors (ICIs) has revolutionized the treatment of advanced NSCLC, there is a growing interest in the role of immunotherapy in early-stage NSCLC. Here, we summarize the rationale for adjuvant immunotherapy, including the postoperative immunosuppressive environment and immunological effects of platinum chemotherapy. Many ongoing clinical trials and the related progress in adjuvant immunotherapy in early-stage resectable NSCLC are discussed. Furthermore, we highlight several unresolved challenges, including markers predictive of treatment benefit, the efficacy of treatment for some oncogene-addicted tumors, the optimal combination therapy, the duration of adjuvant immunotherapy, and optimal selection between neoadjuvant and adjuvant immunotherapy. Early findings in some clinical trials are promising, and updated overall survival results will be useful for validating the current role of adjuvant immunotherapy, particularly in the context of perioperative strategy.

Personalized management of anastomotic leak after surgery for esophageal carcinoma.

OBJECTIVE: To summarize the management of anastomotic leak following surgery for esophageal carcinoma. METHODS: The medical records of the patients developing digestive tract leak after surgery for esophageal carcinoma in our hospital from January 2003 to March 2011 were retrospectively analyzed. RESULTS: A total of 36 patients were included, in whom 13 developed cervical anastomotic leak, 18 had intra-thoracic anastomotic leak, and 5 had intra-thoracic gastric necrosis. Of these patients, 7 were treated with resurgery, 6 with esophageal stent implantation, and 23 with conservative treatment. Treatment lasted for 5 to 181 days, averagely 47.0 +/- 31.9 days. After management, 9 patients died (25.0%). Among seven patients with resurgery, four had deceased, two were cured, and one developed leak again and was switched to conservative treatment until discharged. All the 6 patients treated with stent implantation were cured. Of the 24 patients receiving conservative treatment (including one switched from resurgery), 18 (75.0%) were cured and 1 was not cured but survived. CONCLUSIONS: Anastomotic leak following surgery for esophageal carcinoma should be treated individually based on the onset time, location, size, and extent of the leakage. Conservative treatment is still a safe and effective method. The efficacy of stent implantation needs further investigation to confirm.

Pathologic complete response after neoadjuvant tislelizumab and chemotherapy for Pancoast tumor: A case report.

A 60-year-old man was hospitalized because of numbness and weakness in the right upper limb. Magnetic resonance imaging revealed a large mass in the right upper lobe invading the right eighth cervical and first thoracic nerve root. Biopsy pathology confirmed primary lung adenocarcinoma with a clinical stage of cT4N0M0 IIIA, negative for anaplastic lymphoma kinase fusion gene and epidermal growth factor receptor mutations but positive for programmed death ligand 1 (3%). Neoadjuvant tislelizumab and chemotherapy were offered to this patient with Pancoast tumor, and tumor shrinkage of 71% was achieved. After the operation, surgical pathology indicated pathologic complete response (pCR). Circulating tumor cells testing was negative after the first adjuvant treatment. In this case, we provide real-world evidence of encouraging pCR with neoadjuvant tislelizumab and chemotherapy for a patient with Pancoast tumor.

[Characterization of volatile organic compounds (VOCs) emission from recycled waste polymethyl methacrylate (PMMA) industry].

In the present study, the main source and characteristics of volatile organic compounds (VOCs) released from recycled waste polymethyl methacrylate (PMMA) in Zhejiang province was analyzed quantitatively and qualitatively. Based on those results, the internationally recognized evaluation model of health risk assessment was adopted in the preliminary assessment of the health risk of those VOCs. The results showed that methanol (MeOH) and methyl methacrylate (MMA) were the main pollutants. Emission from part of enterprises exceeded the national standards in MeOH and MMA, especially for fugitive emissions of odor concentration. Moreover, health risk value of fugitive emissions was very low and did not cause harm to human health. Additionally, to better accelerate industrial upgrading and environment pollution treatment, the emission limits of MeOH, MMA and odor concentration were suggested and the values of existing/new (fugitive emission) limitations were 45/30 (0.5) mg x m(-3), 30/25 (0.5) mg x m(-3), and 1000/800 (20), respectively.