Fethullah Karabiber

Major satellite repeat RNA stabilize heterochromatin retention of Suv39h enzymes by RNA-nucleosome association and RNA:DNA hybrid formation

The Suv39h1 and Suv39h2 histone lysine methyltransferases are hallmark enzymes at mammalian heterochromatin. We show here that the mouse Suv39h2 enzyme differs from Suv39h1 by containing an N-terminal basic domain that facilitates retention at mitotic chromatin and provides an additional affinity for major satellite repeat RNA. To analyze an RNA-dependent interaction with chromatin, we purified native nucleosomes from mouse ES cells and detect that Suv39h1 and Suv39h2 exclusively associate with poly-nucleosomes. This association was attenuated upon RNaseH incubation and entirely lost upon RNaseA digestion of native chromatin. Major satellite repeat transcripts remain chromatin-associated and have a secondary structure that favors RNA:DNA hybrid formation. Together, these data reveal an RNA-mediated mechanism for the stable chromatin interaction of the Suv39h KMT and suggest a function for major satellite non-coding RNA in the organization of an RNA-nucleosome scaffold as the underlying structure of mouse heterochromatin. DOI: http://dx.doi.org/10.7554/eLife.25293.001

Performance analysis of Lab2000HL color space for background subtraction

Background subtraction techniques are commonly used to identify moving objects in computer vision applications. This is still a challenging problem, especially when there is a non-stationary background such as a waving sea, or in the case where camera oscillations exist, or when videos have non-stationary backgrounds because of sudden changes in lightning. One of the most significant sub-tasks of a generic background subtraction technique is the background modeling step, which determines how background will be represented. A wide range of the literature is upon development of statistical models for background modeling. Especially, Gaussian Mixture Model (GMM) is a basic method. In this method, values of each pixels features with respect to time are represented with a few normal distributions. The problem with which features pixels will be represented is an important research topic. Recent studies involve applications using different color space with both pixel and region based features. In this study, in addition to color spaces used in literature, new color space which have linear hue band and named as Lab2000HL is aimed to test. The segmentation of foreground/background performance is measured with average precision rate. Nine different videos from I2R dataset having non-static background examples are used as test dataset.

SHAPE analysis of small RNAs and riboswitches.

We describe structural analysis of small RNAs by SHAPE chemical probing. RNAs are treated with 1-methyl-7-nitroisatoic anhydride, a reagent that detects local nucleotide flexibility; and N-methylisatoic anhydride and 1-methyl-6-nitroisatoic anhydride, reagents which together detect higher-order and noncanonical interactions. Chemical adducts are quantified as stops during reverse transcriptase-mediated primer extension. Probing information can be used to infer conformational changes and ligand binding and to develop highly accurate models of RNA secondary structures.

An experimental evaluation of electrical skin conductivity changes in postmortem interval and its assessment for time of death estimation

In forensic medicine, estimation of the time of death (ToD) is one of the most important and challenging medico-legal problems. Despite the partial accomplishments in ToD estimations to date, the error margin of ToD estimation is still too large. In this study, electrical conductivity changes were experimentally investigated in the postmortem interval in human cases. Electrical conductivity measurements give some promising clues about the postmortem interval. A living human has a natural electrical conductivity; in the postmortem interval, intracellular fluids gradually leak out of cells. These leaked fluids combine with extra-cellular fluids in tissues and since both fluids are electrolytic, intracellular fluids help increase conductivity. Thus, the level of electrical conductivity is expected to increase with increased time after death. In this study, electrical conductivity tests were applied for six hours. The electrical conductivity of the cases exponentially increased during the tested time period, indicating a positive relationship between electrical conductivity and the postmortem interval.

Face recognition on mobile environment images using appearance based methods

In this paper, we present face recognition systems, which are performed by using appearance based methods on mobile environment face images, and their comparative performance analysis. In proposed systems, face detection process is performed by using Haar-like features and cascade classifiers on mobile environment face images. Color space transformation, dimensional normalization and histogram equalization operations are performed on detected face images as pre-processing steps. Principal Component Analysis, Fishers Linear Discriminant Analysis and Local Binary Pattern Histograms methods are used to extract facial features. K-nearest neighbor classifier is employed for the performance analysis of implemented methods. Accuracy, precision, recall and F-measure values are measured and compared in performance evaluations of selected facial recognition methods on various dimensionally normalized face images. Experimental results obtained using MOBIO face database show that Local Binary Pattern Histograms method has high success rates on mobile environment images.

Evaluation of I-vector and GMM based speaker verification systems for forensic application

In addition to visual and auditory evaluation, forensic audio experts use automatic speaker verification systems to analyze forensic voice evidence. Gaussian Mixture Model is widely used in such systems. Recently, the usage of i-vector, derived from factor analysis, has gained popularity. In this work, performance comparison of Gaussian Mixture Model and i-vector methods on a speaker verification system is conducted on short recordings of speech. The experimental results revealed that the system that uses i-vector produces a much lower equal error rate.

Implementation of data generator for process mining applications

Process Mining is a new research area that tries to answer the business world needs such as improvement of productivity and efficiency. The aim of process mining is to discover, monitor and improve business processes. Starting point of process mining is event logs which contain information about every step of a business process. With the help of Process Mining techniques, we can analyze the event logs to discover, monitor and improve business processes. The aim of this study is to develop a simulator which generates data that will lead to the development and analysis of business processes. A web system has been created under this study that would work with various business process models and bring out significant data from such models. More productive and efficient processes may be developed using the system. The system consists of three modules. The first module allows the users create a process model. In this module, the users can create their own business model using the interface or can choose among the other models already designed. In the second module, the simulation of the team based data can be performed using the process models already created. In the third module, which is named as “analysis”, some meaningful information such as the time, source and cost can be extracted using generated data. Consequently, the process can be improved by examining the extracted information.