Although additional studies are needed, our findings could provide information regarding the genetic aetiology of IDD in the Korean population.

Located on chromosome 10q22-q23, the human neuregulin3 (NRG3) is considered to be a strong positional and functional candidate gene for schizophrenia pathogenesis. Several case-control studies examining the association of polymorphisms in NRG3 with schizophrenia and/or related traits such as delusion have been reported recently in cohorts of Han Chinese, Ashkenazi Jews, Australians and white Americans of Western European ancestry. Thus, this study aimed to comprehensively investigate the association of NRG3 genetic variations with the risk of schizophrenia and smooth pursuit eye movement (SPEM) abnormality in a Korean population. Using TaqMan assay, six single-nucleotide polymorphisms (SNPs) in the intronic region of NRG3 were genotyped and two major haplotypes were identified in 435 patients with schizophrenia as cases and 393 unrelated healthy individuals as controls. A total of 113 schizophrenia patients underwent an eye tracking task, and degree of SPEM abnormality was measured using the logarithmic values of the signal/noise (Ln S/N) ratio. Differences in frequency distributions were analyzed using logistic and regression models following various modes of genetic inheritance and controlling for age and sex as covariates. Subsequent analysis revealed that the frequency distributions of NRG3 polymorphisms and haplotypes were similar between schizophrenia patients and healthy controls of Korean ethnicity. Furthermore, no significant differences were observed between the genetic variants tested for SPEM abnormality. By elucidating a lack of association in a Korean population, findings from this study may contribute to the understanding of the genetic etiology focusing on the role of NRG3 in schizophrenia pathogenesis.

Machine learning-based time-series forecasting has recently been intensively studied. Deep learning (DL), specifically deep neural networks (DNN) and long short-term memory (LSTM), are the popular approaches for this purpose. However, these methods have several problems. First, DNN needs a lot of data to avoid over-fitting. Without sufficient data, the model cannot be generalized so it may not be good for unseen data. Second, impaired data affect forecasting accuracy. In general, one trains a model assuming that normal data enters the input. However, when anomalous data enters the input, the forecasting accuracy of the model may decrease substantially, which emphasizes the importance of data integrity. This paper focuses on these two problems. In time-series forecasting, especially for photovoltaic (PV) forecasting, data from solar power plants are not sufficient. As solar panels are newly installed, a sufficiently long period of data cannot be obtained. We also find that many solar power plants may contain a substantial amount of anomalous data, e.g., 30%. In this regard, we propose a data preprocessing technique leveraging convolutional autoencoder and principal component analysis (PCA) to use insufficient data with a high rate of anomaly. We compare the performance of the PV forecasting model after applying the proposed anomaly detection in constructing a virtual power plant (VPP). Extensive experiments with 2517 PV sites in the Republic of Korea, which are used for VPP construction, confirm that the proposed technique can filter out anomaly PV sites with very high accuracy, e.g., 99%, which in turn contributes to reducing the forecasting error by 23%.

No abstract