Lisansüstü Eğitim Enstitüsü

Permanent URI for this communityhttps://hdl.handle.net/20.500.12469/77

Browse

Browsing Lisansüstü Eğitim Enstitüsü by Scopus Q "Q2"

Filter results by typing the first few letters
Now showing 1 - 2 of 2
  • Thumbnail Image
    Article
    Citation - WoS: 22
    Citation - Scopus: 34
    A Hybrid Deep Learning Framework for Unsupervised Anomaly Detection in Multivariate Spatio-Temporal Data
    (Mdpi, 2020) Karadayı, Yıldız; Aydın, Mehmet Nafiz; Öğrenci, Arif Selçuk
    Multivariate time-series data with a contextual spatial attribute have extensive use for finding anomalous patterns in a wide variety of application domains such as earth science, hurricane tracking, fraud, and disease outbreak detection. In most settings, spatial context is often expressed in terms of ZIP code or region coordinates such as latitude and longitude. However, traditional anomaly detection techniques cannot handle more than one contextual attribute in a unified way. In this paper, a new hybrid approach based on deep learning is proposed to solve the anomaly detection problem in multivariate spatio-temporal dataset. It works under the assumption that no prior knowledge about the dataset and anomalies are available. The architecture of the proposed hybrid framework is based on an autoencoder scheme, and it is more efficient in extracting features from the spatio-temporal multivariate datasets compared to the traditional spatio-temporal anomaly detection techniques. We conducted extensive experiments using buoy data of 2005 from National Data Buoy Center and Hurricane Katrina as ground truth. Experiments demonstrate that the proposed model achieves more than 10% improvement in accuracy over the methods used in the comparison where our model jointly processes the spatial and temporal dimensions of the contextual data to extract features for anomaly detection.
  • Thumbnail Image
    Doctoral Thesis
    Citation - WoS: 7
    Citation - Scopus: 7
    Screening of Novel and Selective Inhibitors for Neuronal Nitric Oxide Synthase (nnos) Via Structure-Based Drug Design Techniques
    (Kadir Has Üniversitesi, 2022) boumezber, sarah; Yelekci, Kemal
    The overproduction of nitric oxide (NO) by neuronal nitric oxide synthase (nNOS) is the main cause of several neurodegenerative diseases such as Alzheimer’s Disease (AD), Parkinson’s Disease (PD), and Multiple Sclerosis (MS). NO is produced in many cell types by three isoforms of NOS (nNOS, iNOS, and eNOS) and has various biological functions, generally, for its significant reactivity with proteins. NOS isoforms share a high sequence and structure similarity, specifically in the active site, which makes the development and design of nNOS inhibitors extremely challenging; mainly, no-selective inhibitors can affect iNOS and eNOS physiological roles. To date, there is no selective inhibitor against nNOS in the market with desirable ADMET (absorption, distribution, metabolism, elimination, and toxicity) properties, and pass the blood-brain barrier (BBB). With improvement of computational drug design techniques and accessibility of the X-ray crystal structures, development of novel drugs became less expensive and faster. Our research benefited from the structure-based drug design approaches to investigate proficient and selective inhibitors against nNOS. After structure-based virtual screening, the selective top-ranked compounds were filtered according to the ADMET prediction; then, the candidates with a high affinity with a suitable ADMET profile were subject to 100 ns molecular dynamics (MD) simulations. The stability through the 100 ns run has been evident for some nominated inhibitors, which are valuable lead compounds that can be optimized to reach the greatest physicochemical properties in addition to the selectivity.