Berat Doğan | Inönü üniversitesi (original) (raw)

Papers by Berat Doğan

bioRxiv (Cold Spring Harbor Laboratory), May 7, 2019

In this study, segmentation of Multiple Sclerosis (MS) lesions from synthetic brain MRI images wa... more In this study, segmentation of Multiple Sclerosis (MS) lesions from synthetic brain MRI images was aimed by using fuzzy clustering algorithms. The performances of fuzzy c-means algorithm and type-2 fuzzy c-means algorithm were compared. After several experiments it was shown that, the type-2 fuzzy c-means algorithm performed better than the standard fuzzy c-means algorithm.

Computers in Biology and Medicine, Mar 1, 2023

Single-Cell RNA sequencing (scRNA-seq) has provided unprecedented opportunities for exploring gen... more Single-Cell RNA sequencing (scRNA-seq) has provided unprecedented opportunities for exploring gene expression and thus uncovering regulatory relationships between genes at the single cell level. However, scRNA-seq relies on isolating cells from tissues. Thus, the spatial context of the regulatory processes is lost. A recent technological innovation, spatial transcriptomics, allows to measure gene expression while preserving spatial information. A first step in the spatial transcriptomic analysis is to identify the cell type which requires a careful selection of cell-specific marker genes. For this purpose, currently scRNA-seq data is used to select limited number of marker genes from among all genes that distinguish cell types from each other. This study proposes scMAGS (single-cell MArker Gene Selection), a new approach for marker gene selection from scRNA-seq data for spatial transcriptomics studies. scMAGS uses a filtering step in which the candidate genes are extracted prior to the marker gene selection step. For the selection of marker genes, cluster validity indices, Silhouette index or Calinski-Harabasz index (for large datasets) are utilized. Experimental results showed that, in comparison to the existing methods, scMAGS is scalable, fast and accurate. Even for the large datasets with millions of cells, scMAGS could find the required number of marker genes in a reasonable amount of time with less memory requirements.

Protein-RNA interactions play a crucial role in posttranscriptional regulation of gene expression... more Protein-RNA interactions play a crucial role in posttranscriptional regulation of gene expression and have diverse functions in various biological processes. Therefore, identification of protein-RNA interactions is quite important. Experimental methods used for this purpose are expensive, time-consuming and labor intensive. Alternatively, machine learning based methods are proposed to detect protein-RNA interactions computationally. In these methods, each protein-RNA pair is represented by a feature vector which is then used to train machine learning methods. Here, in this study, we also proposed an alternative method to form a feature vector for each protein-RNA pair. Compared to the existing methods, the proposed method creates low-dimensional feature vectors which in turn decreases the overall computational time required to train and test the machine learning methods. Moreover, the proposed method does not make any concession on the classification performance.

Applied Soft Computing, Nov 1, 2016

Archives of Gynecology and Obstetrics

Journal of Obstetrics and Gynaecology Research, 2022

AimThe aim of this study was to determine whether there was a difference in placental metabolite ... more AimThe aim of this study was to determine whether there was a difference in placental metabolite profiles between patients with fetal growth restriction (FGR) and healthy controls.MethodsThe study included 10 patients with FGR diagnosis with 14 healthy controls with both matched maternal age and body mass index. 1H HR‐MAS NMR spectroscopy data obtained from placental tissue samples of patients with FGR and healthy control group were analyzed with bioinformatics methods. The obtained results of metabolite levels were further validated with the internal standard (IS) quantification method.ResultsPrincipal component analysis (PCA) and the partial least squares discriminant analysis (PLS‐DA) score plots obtained with the multivariate statistical analysis of preprocessed spectral data shows a separation between the samples from patients with FGR and healthy controls. Bioinformatics analysis results suggest that the placental levels of lactate, glutamine, glycerophosphocholine, phosphocho...

This study proposes a new single-solution based metaheuristic, namely the Vortex Search algorithm... more This study proposes a new single-solution based metaheuristic, namely the Vortex Search algorithm (VS), for fuzzy clustering of ECG beats. The newly proposed metaheuristic is quite simple and highly competitive when compared to the population-based metaheuristics. In order to study the perfor- mance of the proposed method a number of experiments are performed over a dataset which is created by using the records selected from MIT-BIH arrhyth- mia database. The selected records includes six type of beats, namely, Normal Beat (N), Premature Ventricular Contraction (PVC), Fusion of Ventricular and Normal Beat (F), Atrial Premature Beat (A), Right Bundle Branch Block Beat (R) and Fusion of Paced and Normal Beat (f). The records are first preprocessed and then four morphological features are extracted for each beat type to form the training and test sets. By using the newly proposed method, fuzzy cluster centers of the training set is found. By using these clusters' centers a supervis...

2019 27th Signal Processing and Communications Applications Conference (SIU), 2019

BJOG: An International Journal of Obstetrics & Gynaecology, 2021

ObjectiveTo evaluate the impact of vaginal microbiota on pregnancy outcomes in women undergoing a... more ObjectiveTo evaluate the impact of vaginal microbiota on pregnancy outcomes in women undergoing assisted reproduction.DesignA prospective cohort study.SettingA university‐based assisted reproductive technology (ART) centre.Population223 women undergoing ART treatment.MethodsPrior to embryo transfer, vaginal samples were collected from the posterior fornix. Vaginal microbiota identification was carried out using next‐generation sequencing and categorised according to the V3–V4 hypervariable region in the 16S rRNA gene region.Main outcome measuresART clinical outcomes (implantation, clinical pregnancy rates and live birth rates).ResultsThe live birth rate in women with community state type (CST)‐I (39%) was higher than that in women with CST‐III (21.5%) but the difference was not statistically significant (P = 0.052). The relative abundance of Lactobacillus was lower in women who failed to become pregnant (NP group) (67.71%) than in women who became pregnant (PR group) (79.72%). Howev...

Balkan Journal of Electrical and Computer Engineering, 2019

The available number of protein sequences rapidly increased with the development of new sequencin... more The available number of protein sequences rapidly increased with the development of new sequencing techniques. This in turn led to an urgent need for the development of new computational methods utilizing these data for the solution of different biological problems. One of these problems is the comparison of protein sequences from different species to reveal their evolutional relationship. Recently, several alignment-free methods proposed for this purpose. Here in this study, we also proposed an alignment-free method for the same purpose. Different from the existing methods, the proposed method not only allows for a pairwise comparison of two protein sequences, but also it allows for a bulk comparison of multiple protein sequences simultaneously. Computational results performed on gold-standard datasets showed that, bulk comparison of multiple sequences is much faster than its pairwise counterpart and the proposed method achieves a performance which is quite competitive with the sta...

Methods in molecular biology (Clifton, N.J.), 2018

CysHis zinc-finger proteins (C2H2-ZFPs) constitute the largest class of human transcription facto... more CysHis zinc-finger proteins (C2H2-ZFPs) constitute the largest class of human transcription factors (TFs) and also the least characterized one. Determining the DNA sequence preferences of C2H2-ZFPs is an important first step toward elucidating their roles in transcriptional regulation. Among the most promising approaches for obtaining the sequence preferences of C2H2-ZFPs are those that combine machine-learning predictions with in vivo binding maps of these proteins. Here, we provide a protocol and guidelines for predicting the DNA-binding preferences of C2H2-ZFPs from their amino acid sequences using a machine learning-based recognition code. This protocol also describes the tools and steps to combine these predictions with ChIP-seq data to remove inaccuracies, identify the zinc-finger domains within each C2H2-ZFP that engage with DNA in vivo, and pinpoint the genomic binding sites of the C2H2-ZFPs.

Systems biology in reproductive medicine, Jan 28, 2018

The purpose of this study was to investigate whether a change in the follicular fluid metabolomic... more The purpose of this study was to investigate whether a change in the follicular fluid metabolomics profile due to endometrioma is identifiable. Twelve women with ovarian endometriosis (aged<40 years, with a body mass index [BMI] of <30 kg/m) and 12 age- and BMI-matched controls (women with infertility purely due to a male factor) underwent ovarian stimulation for intracytoplasmic sperm injection (ICSI). Follicular fluid samples were collected from both of groups at the time of oocyte retrieval for ICSI. Next, nuclear magnetic resonance (NMR) spectroscopy was performed for the collected follicular fluids. The metabolic compositions of the follicular fluids were then compared using univariate and multivariate statistical analyses of NMR data. Univariate and multivariate statistical analyses of NMR data showed that the metabolomic profiles of the follicular fluids obtained from the women with ovarian endometriosis were distinctly different from those obtained from the control gro...

Applied Soft Computing, 2016

International Journal of Artificial Intelligence & Applications, 2016

The Vortex Search (VS) algorithm is one of the recently proposed metaheuristic algorithms which w... more The Vortex Search (VS) algorithm is one of the recently proposed metaheuristic algorithms which was inspired from the vortical flow of the stirred fluids. Although the VS algorithm is shown to be a good candidate for the solution of certain optimization problems, it also has some drawbacks. In the VS algorithm, candidate solutions are generated around the current best solution by using a Gaussian distribution at each iteration pass. This provides simplicity to the algorithm but it also leads to some problems along. Especially, for the functions those have a number of local minimum points, to select a single point to generate candidate solutions leads the algorithm to being trapped into a local minimum point. Due to the adaptive step-size adjustment scheme used in the VS algorithm, the locality of the created candidate solutions is increased at each iteration pass. Therefore, if the algorithm cannot escape a local point as quickly as possible, it becomes much more difficult for the algorithm to escape from that point in the latter iterations. In this study, a modified Vortex Search algorithm (MVS) is proposed to overcome above mentioned drawback of the existing VS algorithm. In the MVS algorithm, the candidate solutions are generated around a number of points at each iteration pass. Computational results showed that with the help of this modification the global search ability of the existing VS algorithm is improved and the MVS algorithm outperformed the existing VS algorithm, PSO2011 and ABC algorithms for the benchmark numerical function set.

International Journal of Machine Learning and Computing, 2015

The energy function of the off-lattice AB model has a number of deep valleys and hills which usua... more The energy function of the off-lattice AB model has a number of deep valleys and hills which usually leads the search algorithms to trap into a local minimum point. Existing studies usually performs algorithmic improvements on the well-known search methods to avoid from these local minimum points. However, these algorithmic improvements further increase the computational time which is not desired for the protein folding problem. In this study, it is aimed to smooth the energy landscape of this energy function and thus, to find a near optimal or optimal configuration without performing algorithmic improvements on the search methods. This is achieved by adding an additional term to the original energy function by which a hydrophobic core is formed and a near optimal and optimal configuration is found easily. In the experiments, a newly proposed optimization algorithm, the Vortex Search (VS) algorithm, is used to minimize both the original and modified energy functions. Experimental results showed that, the modified energy function helps the VS algorithm to find the desired configurations much more easier than the original function when the maximum number of iterations is kept equal for both cases.

2015 23nd Signal Processing and Communications Applications Conference (SIU), 2015

Applied Soft Computing, 2015

A new state space representation of the protein folding problem in 2D-HP model is proposed for th... more A new state space representation of the protein folding problem in 2D-HP model is proposed for the use of reinforcement learning methods.The proposed state space representation reduces the dependency of the size of the state-action space to the amino acid sequence length.The proposed state space representation also provides an actual learning for an agent. Thus, at the end of a learning process an agent could find the optimum fold of any sequence of a certain length, which is not the case in the existing reinforcement learning methods.By using the Ant-Q algorithm (an ant based reinforcement learning method), optimum fold of a protein sequence is found rapidly when compared to the standard Q-learning algorithm. In this study, a new state space representation of the protein folding problem for the use of reinforcement learning methods is proposed. In the existing studies, the way of defining the state-action space prevents the agent to learn the state space for any amino-acid sequence, but rather, the defined state-action space is valid for only a particular amino-acid sequence. Moreover, in the existing methods, the size of the state space is strictly depends on the amino-acid sequence length. The newly proposed state-action space reduces this dependency and allows the agent to find the optimal fold of any sequence of a certain length. Additionally, by utilizing an ant based reinforcement learning algorithm, the Ant-Q algorithm, optimum fold of a protein is found rapidly when compared to the standard Q-learning algorithm. Experiments showed that, the new state-action space with the ant based reinforcement learning method is much more suited for the protein folding problem in two dimensional lattice model.

International Journal of Bioscience, Biochemistry and Bioinformatics, 2013