Characterizing driving styles of human drivers using vehicle sensor data, e.g., GPS, is an interesting research problem and an important real-world requirement from automotive industries. A good representation of driving features can be highly valuable for autonomous driving, auto insurance, and many other application scenarios. However, traditional methods mainly rely on handcrafted features, which limit machine learning algorithms to achieve a better performance. In this paper, we propose a novel deep learning solution to this problem, which could be the first attempt of extending deep learning to driving behavior analysis based on GPS data. The proposed approach can effectively extract high level and interpretable features describing complex driving patterns. It also requires significantly less human experience and work. The power of the learned driving style representations are validated through the driver identification problem using a large real dataset.

No abstract

Objective: Human uveal melanoma (UM) is a common ocular malignant tumor with a high risk of metastasis. Emerging evidence indicates that long non-coding RNAs (lncRNAs) are correlated with the development of UM. Here, we aimed to determine the biological significance of lncRNA growth arrest-specific transcript 5 (GAS5) in UM. Methods: The expression levels of GAS5 and microRNA-21 (miR-21) in UM tissues and cells were detected by qRT-PCR analysis. CCK-8 assay was performed to investigate the viability of UM cells after cell transfections, and the migration and invasion of UM cells were determined by transwell assay. The protein expression levels were detected by Western blot assay. The relationship between miR-21 and GAS5 in UM cells was confirmed by bioinformatics prediction and luciferase report assay. Results: Our experiments demonstrated that GAS5 was markedly downregulated in UM cells and clinical specimens. Overexpression of GAS5 inhibited, whereas knockdown of GAS5 promoted the viability, migration, and invasion of UM cells. The epithelial-tomesenchymal transition (EMT) process of UM cells was also suppressed by upregulating of GAS5 and enhanced by downregulating of GAS5. Additionally, as a competitive endogenous RNA (ceRNA), GAS5 directly binded to the oncogenic miR-21 in UM cells, and overexpression of miR-21 attenuated the EMT-suppressing effect of GAS5. Conclusion: Taken together, our findings suggest that GAS5/miR-21 axis is implicated in the pathogenesis of UM and might serve as a potential therapeutic target.

Background/Objective Although lots of long non-coding RNAs (lncRNAs) have been demonstrated to be involved in carcinogenesis, the functions of numerous of lncRNAs remain unknown. Bioinformatics online database showed that lncRNA LOC100132707 was highly expressed in metastatic melanoma tissues, and its expression predicted a lower overall survival rate in melanoma patients. However, LOC100132707 function in uveal melanoma (UM) progression still remains unclear. In the present study, we aimed to elucidate the role and molecular mechanisms underlying LOC100132707 in UM. Methods RT-PCR was used to detect the levels of LOC100132707 in UM cells. Cell migration, invasion and tumorigenesis were tested by using the transwell chamber assay and in vivo assay. Results LOC100132707 expression in metastatic UM cell line MM28 was significantly higher than that of the non-metastatic UM cell lines, MP38, MP46 and MP65, as well as the expressions of LOC100132707-related genes, including XRN1, PARP14, JAK2, DDX60, BUB1 and SAMD9L. LOC100132707 downregulation significantly repressed cell migration and invasion abilities, whereas overexpressing JAK2 rescued these effects. Consistently, upregulation of LOC100132707 induced significant increases in cell migration and invasion abilities via upregulating JAK2. In addition, silencing of LOC100132707 significantly repressed the in vivo tumor formation ability in UM cells. Conclusion This study reveals that silence of LOC100132707 represses the migration of UM via downregulating JAK2. The LOC100132707/JAK2 axis might serve as a potent target for the prevention and treatment of UM metastasis.

No abstract