Computer Engineering / Bilgisayar Mühendisliği
Permanent URI for this collectionhttps://hdl.handle.net/11147/10
Browse
5 results
Search Results
Article Citation - WoS: 7Citation - Scopus: 5A Machine Learning Approach for Microrna Precursor Prediction in Retro-Transcribing Virus Genomes(Informationsmanagement in der Biotechnologie e.V. (IMBio e.V.), 2016) Saçar Demirci, Müşerref Duygu; Toprak, Mustafa; Allmer, JensIdentification of microRNA (miRNA) precursors has seen increased efforts in recent years. The difficulty in experimental detection of pre-miRNAs increased the usage of computational approaches. Most of these approaches rely on machine learning especially classification. In order to achieve successful classification, many parameters need to be considered such as data quality, choice of classifier settings, and feature selection. For the latter one, we developed a distributed genetic algorithm on HTCondor to perform feature selection. Moreover, we employed two widely used classification algorithms libSVM and random forest with different settings to analyze the influence on the overall classification performance. In this study we analyzed 5 human retro virus genomes; Human endogenous retrovirus K113, Hepatitis B virus (strain ayw), Human T lymphotropic virus 1, Human T lymphotropic virus 2, Human immunodeficiency virus 2, and Human immunodeficiency virus 1. We then predicted pre-miRNAs by using the information from known virus and human pre-miRNAs. Our results indicate that these viruses produce novel unknown miRNA precursors which warrant further experimental validation.Article Citation - WoS: 30Citation - Scopus: 34Gddom: an Online Tool for Calculation of Dominant Marker Gene Diversity(Springer Verlag, 2017) Abuzayed, Mazen; El-Dabba, Nourhan; Frary, Anne; Doğanlar, SamiGene diversity (GD), also called polymorphism information content, is a commonly used measure of molecular marker polymorphism. Calculation of GD for dominant markers such as AFLP, RAPD, and multilocus SSRs is valuable for researchers. To meet this need, we developed a free online computer program, GDdom, which provides easy, quick, and accurate calculation of dominant marker GD with a commonly used formula. Results are presented in tabular form for quick interpretation. © 2016, Springer Science+Business Media New York.Book Part Citation - WoS: 299Citation - Scopus: 406Introduction To Machine Learning(Humana Press, 2014) Baştanlar, Yalın; Özuysal, MustafaThe machine learning field, which can be briefly defined as enabling computers make successful predictions using past experiences, has exhibited an impressive development recently with the help of the rapid increase in the storage capacity and processing power of computers. Together with many other disciplines, machine learning methods have been widely employed in bioinformatics. The difficulties and cost of biological analyses have led to the development of sophisticated machine learning approaches for this application area. In this chapter, we first review the fundamental concepts of machine learning such as feature assessment, unsupervised versus supervised learning and types of classification. Then, we point out the main issues of designing machine learning experiments and their performance evaluation. Finally, we introduce some supervised learning methods.Article Citation - WoS: 24Citation - Scopus: 23The Effect of Age, Menopausal State, and Breast Density on 18f-Fdg Uptake in Normal Glandular Breast Tissue(Society of Nuclear Medicine and Molecular Imaging, 2010) Mavi, Ayşe; Çermik, Tevfik F.; Urhan, Muammer; Püskülcü, Halis; Basu, Sandip; Cucchiara, Andrew J.; Yu, Jian Q.; Alavi, AbassTheoretically, the degree of 18F-FDG uptake in the glandular tissues of the normal breast can affect the detection of breast cancer. The aim of this prospective study was to investigate relationships among age, menopausal state, and breast density and determine whether they affect 18F-FDG uptake in normal glandular breast tissue. Methods: Among 250 newly diagnosed breast cancer patients, 149 patients (mean age ± SD, 50.9 ± 9.70 y; range, 32-77 y) were analyzed because they had normal contralateral breasts confirmed by MRI, mammography, and 18F-FDG PET examinations. PET images were acquired 60 ± 2 min after the administration of 18F-FDG (5.2 MBq/kg of body weight). The maximum and average standardized uptake value (SUVmax and SUVavg, respectively) of 18F-FDG were calculated in the normal breast. Patients were divided into groups according to qualitative breast density and menopausal state. Descriptive statistics and 2-factorial analysis of covariance were used to assess the effects of qualitative breast density, menopausal state, and age on SUVmax and SUVavg. Pearson χ2 was used to test the relationship between menopausal state and qualitative breast density. Results: The average age of patients with nondense breasts was significantly higher than that of patients with dense breasts (P < 0.01). Also, breast density related to menopausal state (P < 0.05). Dense breasts had an average SUVmax of 1.243 and mean SUVavg of 0.694, whereas nondense breasts had a mean SUVmax of 0.997 and mean SUVavg of 0.592. Analysis of covariance indicated that density and the linear effect of age were significant with regard to both SUVmax and SUVavg. After removing the linear effect of age, menopausal state had no effect on SUVmax and SUVavg. Conclusion: 18F-FDG uptake significantly decreases as age increases and breast density decreases. Age and qualitative breast density are independent factors and significantly affect 18F-FDG uptake for both SUVmax and SUVavg. Menopausal state had no effect on SUVmax and SUVavg. Copyright © 2010 by the Society of Nuclear Medicine, Inc.Article Citation - WoS: 72Citation - Scopus: 75The Effects of Estrogen, Progesterone, and C-Erbb Receptor States on 18f-Fdg Uptake of Primary Breast Cancer Lesions(Society of Nuclear Medicine and Molecular Imaging, 2007) Mavi, Ayşe; Çermik, Tevfik F.; Urhan, Muammer; Püskülcü, Halis; Basu, Sandip; Yu, Jian Q.; Zhuang, Hongming; Czerniecki, Brian; Alavi, AbassThe purpose of this prospective study was to investigate whether correlations exist between 18F-FDG uptake of primary breast cancer lesions and predictive and prognostic factors such as estrogen receptor (ER), progesterone receptor (PR), and C-erbB-2 receptor (C-erbB-2R) states. Methods: Before undergoing partial or total mastectomy, 213 patients with newly diagnosed breast cancer underwent 18F-FDG PET (5.2 MBq/kg of body weight). The maximum standardized uptake value (SUV) of the primary lesion was measured in each patient. Standard immunohistochemistry was performed on a surgical specimen of the cancer lesion to characterize the receptor state of the tumor cells. Pearson χ2 tests were performed on the cross-tables of different receptor states to test any association that may exist among ER, PR, and C-erbB-2R. Maximum SUV measurements for different receptor states were compared using factorial ANOVA in a completely random design. Results: After exclusion of certain lesions, 118 lesions were analyzed for this study. The mean maximum SUVs of ER-positive and ER-negative lesions were 3.03 ± 0.26 and 5.64 ± 0.75, whereas those of PR were 3.24 ± 0.29 and 4.89 ± 0.67, respectively, and those of C-erbB-2R were 4.64 ± 0.70 and 3.70 ± 0.35, respectively, χ2 tests for ER and PR showed that if one is positive then the other tends to be positive as well (χ2 = 71.054, P < 0.01). For ER and C-erbB-2R states, if ER is positive, C-erbB-2R will more likely be negative (χ2 = 13.026, P < 0.01). No relationship was detected between PR and C-erbB-2R states (χ2 = 3.695, P > 0.05). ANOVAs showed that PR state alone (F = 0.095, P > 0.05) and C-erbB-2R state alone (F = 0.097, P > 0.05) had no effect on 18F-FDG uptake but ER state alone had an effect (F = 9.126, P < 0.01). ER and PR being together had no additional effect on 18F-FDG uptake. Our study also demonstrated that interactions exist between ER and C-erbB-2R state and between PR and C-erbB-2R state. Conclusion: SUV measurements may provide valuable information about the state of ER, PR, and C-erbB-2R and the associated glucose metabolism as measured by 18F-FDG uptake of the primary breast cancer lesions. Such an association may be of importance to treatment planning and outcome in these patients.