No abstract

Background We aimed to develop and validate a real-time deep convolutional neural networks (DCNNs) system for detecting early gastric cancer (EGC). Methods All 45,240 endoscopic images from 1364 patients were divided into a training dataset (35823 images from 1085 patients) and a validation dataset (9417 images from 279 patients). Another 1514 images from three other hospitals were used as external validation. We compared the diagnostic performance of the DCNN system with endoscopists, and then evaluated the performance of endoscopists with or without referring to the system. Thereafter, we evaluated the diagnostic ability of the DCNN system in video streams. The accuracy, sensitivity, specificity, positive predictive value, negative predictive value and Cohen's kappa coefficient were measured to assess the detection performance. Finding The DCNN system showed good performance in EGC detection in validation datasets, with accuracy (85.1%–91.2%), sensitivity (85.9%–95.5%), specificity (81.7%–90.3%), and AUC (0.887–0.940). The DCNN system showed better diagnostic performance than endoscopists and improved the performance of endoscopists. The DCNN system was able to process oesophagogastroduodenoscopy (OGD) video streams to detect EGC lesions in real time. Interpretation We developed a real-time DCNN system for EGC detection with high accuracy and stability. Multicentre prospective validation is needed to acquire high-level evidence for its clinical application. Funding This work was supported by the National Natural Science Foundation of China (grant nos. 81672935 and 81871947), Jiangsu Clinical Medical Center of Digestive System Diseases and Gastrointestinal Cancer (grant no. YXZXB2016002), and Nanjing Science and Technology Development Foundation (grant no. 2017sb332019).

Background Chloroplast and endoplasmic reticulum (ER)-localized fatty acid (FA) transporters have been reported to play important roles in oil (mainly triacylglycerols, TAG) biosynthesis. However, whether these FA transporters synergistically contribute to lipid accumulation, and their effect on lipid metabolism in microalgae are unknown. Results Here, we co-overexpressed two chloroplast-localized FA exporters (FAX1 and FAX2) and one ER-localized FA transporter (ABCA2) in Chlamydomonas. Under standard growth conditions, FAX1/FAX2/ABCA2 over-expression lines (OE) accumulated up to twofold more TAG than the parental strain UVM4, and the total amounts of major polyunsaturated FAs (PUFA) in TAG increased by 4.7-fold. In parallel, the total FA contents and major membrane lipids in FAX1/FAX2/ABCA2-OE also significantly increased compared with those in the control lines. Additionally, the total accumulation contribution ratio of PUFA, to total FA and TAG synthesis in FAX1/FAX2/ABCA2-OE, was 54% and 40% higher than that in UVM4, respectively. Consistently, the expression levels of genes directly involved in TAG synthesis, such as type-II diacylglycerol acyltransferases (DGTT1, DGTT3 and DGTT5), and phospholipid:diacylglycerol acyltransferase 1 (PDAT1), significantly increased, and the expression of PGD1 (MGDG-specific lipase) was upregulated in FAX1/FAX2/ABCA2-OE compared to UVM4. Conclusion These results indicate that the increased expression of FAX1/FAX2/ABCA2 has an additive effect on enhancing TAG, total FA and membrane lipid accumulation and accelerates the PUFA remobilization from membrane lipids to TAG by fine-tuning the key genes involved in lipid metabolism under standard growth conditions. Overall, FAX1/FAX2/ABCA2-OE shows better traits for lipid accumulation than the parental line and previously reported individual FA transporter-OE. Our study provides a potential useful strategy to increase the production of FA-derived energy-rich and value-added compounds in microalgae.

No abstract