Yazar "Altuntas, Volkan" seçeneğine göre listele
Listeleniyor 1 - 16 / 16
Sayfa Başına Sonuç
Sıralama seçenekleri
Öğe APPLICATION OF DEVELOPING CLOTHING RECOMMENDATION SYSTEM WITH ARTIFICIAL INTELLIGENCE TECHNIQUES(2024) Özbek, Ahmet; Altuntas, Volkan; Erdogan, NaıleThis study aims to develop a clothing recommendation application for users who possess a large number of clothes but have limited time due to their intense work tempo. This application aims to assist them in using their clothes effectively, reducing the time spent on selecting outfits, and dressing fashionably. In the process of developing this application, firstly, the criteria influencing the user's clothing preferences were determined. Subsequently, a wardrobe dataset was created based on the established criteria. Following this, methods for suggesting clothes were explored. As a result of the research, it was decided to utilize association rule analysis, multidimensional clothing representation coding, and weighted L1 distance methods for clothing recommendation in this study. In the application phase, experiments were conducted using the dataset associated with the chosen methods. It has been determined that the application developed in this study gives successful results in suggesting clothes suitable for user preferences.Öğe Biyolojik Protein Fonksiyon Tahmin İşleminde Orange Veri Analizi Aracının Kullanımıyla Makine Öğrenmesi Algoritmalarının Performanslarının Değerlendirilmesi(2024) Akman, Ceren; Altuntas, Volkanİnsan vücudu ilk günden bugüne kadar olan bütün süreçlerde işleyiş açısından merak uyandıran bir mekanizma olmuştur. İçerisinde barındırdığı hücrelerle bu hücrelerin kendi içlerinde barındırdıkları moleküllerle ve işleyişlerle yaşamsal döngü devam etmiştir ve devam etmektedir. Bu yaşamsal döngünün devam etmesi için moleküllerin işleyiş şekillerinin anlaşılmasının yaşamsal faaliyetlerin çözümlenmesinde önemli etkisi olduğu kanısına varılmıştır. Bu çalışma kapsamında yapılan çalışmalar incelendiğinde insan vücudu için karmaşık bir yapıya sahip olan moleküllerin işleyişinin büyük bir öneme sahip olduğu kararına varılmıştır. Böylelikle bu çalışma da büyük bir öneme sahip olan karmaşık yapılı protein molekülü ele alınarak biyoloji tarafından bakıldığında biyolojik süreç, moleküler işlev ve hücresel bileşen açısından fonksiyon tahmin işleminin gerçekleştirilebilmesi ve bunun için k- en yakın komşuluk, sinir ağı ve rastgele orman yöntemlerinin veri görselleştirme ve veri analiz aşamasında kullanılabilen Orange editörü vasıtasıyla modellerin geliştirilmesi sağlanmış olup performans değerlendirilmesi yapılmıştır. Yapılan değerlendirmeler sonucunda k-en yakın komşuluk modelinin kullanılan veri setleri üzerinde en az %88 üzerinde başarı sağladığı tespit edilmiştir.Öğe DeepTFBS: Transkripsiyon Faktörü Bağlanma Bölgeleri Tahmini İçin Derin Öğrenme Yöntemleri Kullanan Hibrit Bir Model(2025) Hatipoğlu, Ayşegül; Altuntas, VolkanMoleküler seviyede genetik verinin oluşum, aktarım ve düzenlenme süreçleri anlaşılması zor karmaşık kombinasyonel süreçlerden oluşmaktadır. Bu süreçlerin temelini oluşturan transkripsiyon faktörleri genetik bilginin DNA'dan RNA'ya kopyalanmasını sağlayarak hücrelerin özellik ve fonksiyonlarını belirlemede kritik rol oynar. Özellikle sinir sistemi gibi karmaşık yapıları kontrol eden transkripsiyon faktörleri, gen ifadesini düzenleyerek hastalık, sağlık gibi durumların belirlenmesinde hayati rol oynarlar. Proteinlerin DNA üzerinde bağlandıkları bölgeler, gen ifadelerinin kritik noktalarını belirler ve hücrelerin çeşitli koşullara uyum sağlamasına katkıda bulunur. Genetik hastalıkların teşhis edilmesi ve tedavi edilmesi süreçleri için önemli bir adım olan transkripsiyon faktörü bağlanma bölgelerinin tahmini amacıyla literatürde çeşitli yöntemler geliştirilmiştir. DNA’nın dizi ve şekil özelliklerinin beraber kullanımıyla başarılı sonuçlar elde edilen çeşitli çalışmalar geliştirilmiştir. Bu çalışmada DNA dizileri ve şekillerine dayalı olarak transkripsiyon faktörü etkileşimlerini belirlemek için farklı derin öğrenme teknolojileri birleştirilerek hibrit bir yöntem önerilmiştir. Çalışmada 165 doğrulanmış CHIP-Seq veri kümesi kullanılmıştır.Öğe Diffusion Alignment Coefficient (DAC): A Novel Similarity Metric for Protein-Protein Interaction Network(Ieee Computer Soc, 2023) Altuntas, VolkanInteraction networks can be used to predict the functions of unknown proteins using known interactions and proteins with known functions. Many graph theory or diffusion-based methods have been proposed, using the assumption that the topological properties of a protein in a network are related to its biological function. Here we seek to improve function prediction by finding more similar neighbors with a new diffusion-based alignment technique to overcome the topological information loss of the node. In this study, we introduce the Diffusion Alignment Coefficient (DAC) algorithm, which combines diffusion, longest common subsequence, and longest common substring techniques to measure the similarity of two nodes in protein interaction networks. As a proof of concept, our experiments, conducted on a real PPI networks S.cerevisiae and Homo Sapiens, demonstrated that our method obtained better results than competitors for MIPS and MSigDB Collections hallmark gene set functional categories. This is the first study to develop a measure of node function similarity using alignment to consider the positions of nodes in protein-protein interaction networks. According to the experimental results, the use of spatial information belonging to the nodes in the network has a positive effect on the detection of more functionally similar neighboring nodes.Öğe Identification of peptides derived from ripened Mihalic cheese and promising properties: in silico and in vitro approaches(Springer, 2025) Altuntas, Seda; Altuntas, VolkanDairy products, particularly fermented ones like cheese, are abundant sources of bioactive peptides (BPs) that hold poten-tial health benefits. This study focuses on investigating the peptide profiles of Mihalic cheese, a traditional Turkish hard-brined cheese, employing both in vitro and in silico approaches. This study compares these two methodologies and assesses the potential of in silico digestion for predicting the bioactive peptide profile of dairy products. The in vitro analysis revealed a significant increase in the number of peptides with lengths of <= 10 amino acids after hydrolysis. While 7.7% of the water-soluble peptide extract (WSPE) fraction contained bioactive peptide sequences comprising more than 20 amino acids, all bioactive peptides detected in the hydrolysate of Mihalic cheese consisted of 7-20 amino acids in length. Furthermore, the in vitro analysis identified 30 peptides with various biological activities such as antioxidant, antimicro-bial, angiotensin-converting enzyme (ACE) inhibitory, DPP-IV inhibitory etc., with beta-CN contributing the most to these bioactive peptides. Despite the potential of in silico methods, there was limited overlap with in vitro findings. While in silico analysis predicted a broader range of bioactive peptides, including 17 ACE-inhibitory peptides, only ten peptides and also only three bioactive peptides were identified by both approaches. While in silico tools are valuable for prediction, they may not fully replicate the complex dynamics of actual enzymatic digestion. This highlights the need to improve results by using more combined in vitro and in silico approaches in studies on the prediction of bioactive peptides in food products.Öğe Improving Long Non-Coding RNA Prediction through Recursive Feature Elimination and XGBoost(Gazi Univ, 2025) Alizada, Freshta; Altuntas, VolkanIn recent years, advancements in high-throughput technologies have uncovered numerous concealed layers known as Non-Coding Ribonucleic Acids (ncRNAs), shifting the protein-centric view of genomes. NcRNAs, previously considered insignificant segments of the genome, are now recognized as essential functional components in prokaryotic and eukaryotic organisms. Long non-coding RNAs (lncRNAs) are a unique category of ncRNAs with 200 nucleotides length, which are instrumental in key biological functions, including cellular differentiation, regulatory mechanisms, and epigenetic modifications. Despite the similarities between lncRNAs and messenger RNAs (mRNAs), there is a fundamental difference: mRNAs encode proteins, whereas lncRNAs do not. This study aims to distinguish these two RNA classes from each other by designing a robust machine learning (ML) pipeline employing Recursive Feature Elimination (RFE) for dimensionality reduction of dataset and XGBoost (XGB) classification model. Whereas previous studies trained and tested machine learning models using the complete set of dataset features, we employ the RFE technique to reduce the number of features, thereby we achieve a more optimal dataset with relevant features. To evaluate the predictive performance of our pipeline, we used error rate, accuracy, precision, recall, and F1-score. Compared to three existing lncRNA identification tools in the literature, our pipeline demonstrated superior prediction accuracy and precision at 93.42% and 94.19% respectively.Öğe Integration of Algorithmic and Local Approaches for Link Prediction: An Analysis on Protein-Protein Interactions and Social Networks(Gazi Univ, 2025) Kadem, Hasibe Candan; Altuntas, VolkanComplex network analysis is applied in various fields such as network-based systems, social media recommendation systems, shopping platforms, and treatment methodologies. In this context, predicting the probability of connection between two nodes has become a focal point. Another significant aspect is the prediction of connections between proteins, especially with the increase in epidemic diseases. Link prediction methods, based on graph structures, aim to predict interactions between two nodes and measure the probability of connection between them. These methods proceed by relying on similarity values and can have multiple approaches, including local, global, and algorithmic. This study has emerged from a combination of algorithmic and local network approaches. Support Vector Machines are employed to predict connections in gene-protein networks and social network structures. Data sets from multiple social media platforms and human protein-protein interaction (PPI) data were utilized. Derived data were created by calculating local index values, including the number of neighbors, Adamic Adar index, Jaccard coefficient, and label values for each node. To enhance success rates, a model was developed that applied the discretization method as a preprocessing technique across all data sets. Machine learning algorithms such as Bayesian Networks, Multilayer Perceptron (MLP), Random Forest, and k-Nearest Neighborhood (kNN) were compared and evaluated. The results indicate that the Twitch dataset, which has the highest number of edges, produced successful outcomes. The contribution of edge numbers in the network structure to performance is highlighted, and it is observed that more successful metric values were obtained for the data with applied discretization.Öğe Integration of Algorithmic and Local Approaches for Link Prediction: An Analysis on Protein-Protein Interactions and Social Networks(2025) Kadem, Hasibe Candan; Altuntas, VolkanComplex network analysis is applied in various fields such as network-based systems, social media recommendation systems, shopping platforms, and treatment methodologies. In this context, predicting the probability of connection between two nodes has become a focal point. Another significant aspect is the prediction of connections between proteins, especially with the increase in epidemic diseases. Link prediction methods, based on graph structures, aim to predict interactions between two nodes and measure the probability of connection between them. These methods proceed by relying on similarity values and can have multiple approaches, including local, global, and algorithmic. This study has emerged from a combination of algorithmic and local network approaches. Support Vector Machines are employed to predict connections in gene-protein networks and social network structures. Data sets from multiple social media platforms and human protein-protein interaction (PPI) data were utilized. Derived data were created by calculating local index values, including the number of neighbors, Adamic Adar index, Jaccard coefficient, and label values for each node. To enhance success rates, a model was developed that applied the discretization method as a preprocessing technique across all data sets. Machine learning algorithms such as Bayesian Networks, Multilayer Perceptron (MLP), Random Forest, and k-Nearest Neighborhood (kNN) were compared and evaluated. The results indicate that the Twitch dataset, which has the highest number of edges, produced successful outcomes. The contribution of edge numbers in the network structure to performance is highlighted, and it is observed that more successful metric values were obtained for the data with applied discretization.Öğe Modelling of Listeria monocytogenes growth and survival in presence of royal jelly, a promising anti-biofilm agent(Vup Food Research Inst, Bratislava, 2020) Altuntaş, Seda; Cinar, Aycan; Altuntas, VolkanRoyal jelly is a natural bee product with well-known antimicrobial properties and prominent clinical uses. The main objective of this study was to reveal the response of L. monocytogenes to royal jelly. Firstly, the influence of royal jelly on the growth kinetic parameters of L. monocytogenes ATCC 7644 was studied with the modified Gompertz mathematical equation. Results indicated that royal jelly retarded the growth of L. monocytogenes by acting mainly on the lag phase duration. Microdilution assay on L. monocytogenes ATCC 7644 performed with two-fold serial dilution method showed that minimal inhibitory concentration (MIC) was 41.67 mg.ml(-1). However, the estimated MIC value modelled with Gompertz survival curve (23.85 mg.ml(-1)) and the validated value (24 mg.ml(-1)) were notably lower than in the micro-dilution assay. It is noteworthy that the predictive microbiology approach has good performance in determining the optimum dose and cost-effective of antimicrobial agents. To our knowledge, biofilm prevention by royal jelly has not yet been studied and a limited number of studies regarded the effects on Listeria monocytogenes. The ability to reduce the formation of biofilm with royal jelly may allow further studies to explore the use of apiculture products in the struggle with L. monocytogenes.Öğe NodeVector: A Novel Network Node Vectorization with Graph Analysis and Deep Learning(Mdpi, 2024) Altuntas, VolkanNetwork node embedding captures structural and relational information of nodes in the network and allows for us to use machine learning algorithms for various prediction tasks on network data that have an inherently complex and disordered structure. Network node embedding should preserve as much information as possible about important network properties where information is stored, such as network structure and node properties, while representing nodes as numerical vectors in a lower-dimensional space than the original higher dimensional space. Superior node embedding algorithms are a powerful tool for machine learning with effective and efficient node representation. Recent research in representation learning has led to significant advances in automating features through unsupervised learning, inspired by advances in natural language processing. Here, we seek to improve the representation quality of node embeddings with a new node vectorization technique that uses network analysis to overcome network-based information loss. In this study, we introduce the NodeVector algorithm, which combines network analysis and neural networks to transfer information from the target network to node embedding. As a proof of concept, our experiments performed on different categories of network datasets showed that our method achieves better results than its competitors for target networks. This is the first study to produce node representation by unsupervised learning using the combination of network analysis and neural networks to consider network data structure. Based on experimental results, the use of network analysis, complex initial node representation, balanced negative sampling, and neural networks has a positive effect on the representation quality of network node embedding.Öğe Physicochemical and microbiological investigation of ballast waters of the ships operating in the Marmara Sea(Elsevier Sci Ltd, 2024) Dobrucali, Erinc; Uyanik, Sinan; Altuntas, Volkan; Yilmaz, Mete; Balci, Muharrem; Sahan, Aybuke Nur; Ucar, DenizBallast water, an essential component of global shipping operations, plays a pivotal role in maintaining vessel stability and load distribution. However, its inadvertent discharge can introduce a myriad of physicochemical and microbiological hazards to marine ecosystems, necessitating rigorous investigation. This study presents a comprehensive analysis of ballast waters from ships operating in the ecologically significant Marmara Sea. Different than previous studies, physicochemical parameters, including pH, heavy metal concentrations (Cr, Fe, Co, Ni, Cu, Zn, and As), total organic carbon, turbidity, total nitrogen, and total phosphorus, were extensively assessed. Furthermore, microbial communities were examined through the identification of bacterial, archaeal, and algal taxa using 16S and 18S rRNA gene amplicon sequence data. Particular attention was given to potential pathogens and harmful algal species, employing advanced techniques to ensure accuracy and comprehensiveness. The findings reveal notable variations in the physicochemical profiles of ballast waters, attributed to diverse geographical origins and operational factors. Furthermore, the microbial analysis identifies a diverse array of species, including pathogenic strains and potentially toxin-producing algae, raising concerns about potential ecological and public health implications. These results underscore the urgent need for improved ballast water management strategies and the implementation of effective treatment technologies to mitigate the adverse effects of ballast water discharge in the Marmara Sea. By shedding light on the intricacies of ballast water composition, this study contributes valuable insights toward safeguarding marine biodiversity and human wellbeing in this ecologically sensitive region.Öğe Prediction of Cancer in DNA Sequences Using Unsupervised Learning Methods(2023) Doğru, Şeyma; Altuntas, VolkanToday, with the development of technology, the decision-making capabilities of machines have also increased. With their high analytical skills, computers can easily catch points and relationships that may escape the human eye. Thanks to these capabilities, machines are also widely used in the field of health. For example, many machine-learning techniques developed on cancer prediction have been successfully applied. Early detection of cancer is crucial to survival. In the early diagnosis of cancer, the rates of drug treatment, chemotherapy, or radiotherapy that the person will be exposed to are significantly reduced and the patient gets through this process with the least amount of wear and tear. Gene Expression Cancer RNA-Seq Dataset was used in this study. This data set includes gene expression values of 5 cancer types (BRCA, KIRC, LUAD, LUSC, UCEC). DNA sequences in the dataset were analyzed using k-means and hierarchical clustering algorithms, which are unsupervised machine learning methods. The aim of the study is to develop a usable machine-learning model for the early detection of cancer at the gene level. Adjusted Rand Index (ARI), Silhouette Score, and Accuracy Metrics were used to evaluate the analysis results. The rand index calculates the similarity between clusters by counting the binaries assigned to clusters. The adjusted Rand Index is a randomly adjusted version of the Rand Index. The silhouette score indicates how well a data point fits within its own set among separated datasets. The accuracy metric is obtained as a percentage of correctly clustered data points divided by all predictions. Different connection methods are used in the hierarchical clustering algorithm. These are 'complete', 'ward', 'average', and 'single'. As a result of the study, the accuracy in the k-means algorithm was 0.990, the Adjusted Rand Index was 0.79, and the Silhouette Score was 0.14. Looking at the hierarchical clustering, ward performed the best of the four linkage methods, with an ARI score of 0.76 and a silhouette score of 0.13. As a result of the study, the accuracy of the hierarchical clustering algorithm was 0.999.Öğe Protein complex detection from protein protein i nteraction networks with machine learning methods(2024) Karakuş, Yasin; Altuntas, VolkanUnderstanding Protein - Protein interaction networks, which show the interactions between proteins involved in tasks that are very important for our organisms such as structural support, storage, signal transduction and defence, provides a better understanding of cellular processes. One of the important studies carried out for this purpose is to try to detect protein complexes from protein - protein interaction networks. Supervised and unsupervised machine learning methods were used to detect protein complexes. It is known that the machine learning methods used produce better performance when more than one method is used together. Based on this knowledge, a method that detects protein complexes from protein-protein interaction networks is proposed in this study. The method first weights protein-protein interaction networks using biological and topological properties of proteins. Then it estimates local and global protein complex core. Then it builds a protein complex detection model using the structural modularity of proteins and the voting regression model. We predict that XGB regression, gaussian process regression, catboost regression and histogram-based gradient boosting regression supervised learning methods can achieve more successful results when used together in the voting regression model. When we compare the success of the model with other models, it has shown the best performance many times among the compared models.Öğe Protein complex detection from protein-protein interaction networks with machine learning methods(Pamukkale Univ, 2024) Karakus, Yasin; Altuntas, VolkanUnderstanding Protein - Protein interaction networks, which show the interactions between proteins involved in tasks that are very important for our organisms such as structural support, storage, signal transduction and defence, provides a better understanding of cellular processes. One of the important studies carried out for this purpose is to try to detect protein complexes from protein - protein interaction networks. Supervised and unsupervised machine learning methods were used to detect protein complexes. It is known that the machine learning methods used produce better performance when more than one method is used together. Based on this knowledge, a method that detects protein complexes from protein-protein interaction networks is proposed in this study. The method first weights protein-protein interaction networks using biological and topological properties of proteins. Then it estimates local and global protein complex core. Then it builds a protein complex detection model using the structural modularity of proteins and the voting regression model. We predict that XGB regression, gaussian process regression, catboost regression and histogram -based gradient boosting regression supervised learning methods can achieve more successful results when used together in the voting regression model. When we compare the success of the model with other models, it has shown the best performance many times among the compared models.Öğe RNA m6A Modifikasyon Bölgelerinin Sınıflandırılması için Öznitelik Çıkarma ve Boyut Azaltma Yöntemlerinin Karşılaştırılması(2025) Nuray, Batuhan; Altuntas, VolkanBu çalışmada RNA’da sıklıkla meydana gelen N6-metiladenozin (m6A) modifikasyon bölgelerinin belirlenmesi ve gelecekte yapılacak çalışmalar için farklı öznitelik çıkarıcılar, öznitelik seçiciler ve boyut düşürme algoritmalarının, K-en yakın komşu sınıflandırma algoritması kullanılarak performanslarının karşılaştırılması amaçlanmıştır. 35 farklı öznitelik çıkarma algoritması ve 9 farklı boyut azaltma ve öznitelik seçici algoritma kullanılarak algoritmaların m6A modifikasyon bölgelerinin tanımlamasındaki performansları değerlendirilmiştir. Yapılan çalışmanın sonunda Nükleotidlerin kimyasal özelliklerini dikkate alarak öznitelik çıkarımı yapan NCP öznitelik çıkarma algoritması ile Ekstra Ağaçlar boyut azaltma yönteminin birlikte kullanılmasının m6A modifikasyon bölgelerinin belirlenmesinde yüksek performans gösterdiği görülmüştür.Öğe Yapay Sinir Ağları Kullanılarak Protein Katlanması Tanıma(2023) Dikici, Sena; Altuntas, VolkanProteinler uzun aminoasit zincirlerinden oluşur ve vücut kimyasını düzenlemekle birlikte hücrelerin yapısı ve aralarındaki iletişim için öneme sahiptir. Bir proteinin hücre bazındaki görevini gerçekleştirebilmesi için, molekülü hücredeki hedefiyle etkileşime girebilecek üç boyutlu yapıya dönüştüren bir bükülme süreci olan katlanma işlemini gerçekleştirmesi gerekir. Sıcaklık, ağır metaller veya kimyasal durumlar gibi etkenler proteinlerin yanlış katlanmasına sebep olabilir. Yanlış katlanan proteinler, vücuttaki görevini yerine getiremez. Alzaymır, kistik fibrozis, deli dana hastalığı gibi hastalıklara sebep olabilir. Protein katlanması tanıma işlemi, biyologlar açısından bir problem olarak değerlendirilir. Literatürde yer alan şablon tabanlı yaklaşımlara karşın yapay sinir ağları, protein katlanması probleminin çözümüne yönelik yüksek başarım gösterir. Yapay sinir ağları, ele alınan problemin çözümü için geniş veri kümelerinde yer alan ve problemin çözümüne katkı sağlayacak bilgi kazancı yüksek özellikleri kullanan bir hesaplama tekniğidir. Bu çalışmada SCOPe 2.06, SCOPe 2.07, SCOPe 2.08 veri setleri kullanılarak şablon tabanlı yaklaşımlardan elde edilen sonuçların yapay sinir ağı yöntemi ile birleştirilerek protein katlanması tanıma işlemi gerçekleştirilmiştir. Gerçekleştirilen deneyler sonucunda yapay sinir ağı yönteminin katkısı ile literatürde yer alan sonuçların iyileştirildiği görülmüştür. Bu çalışma ile biyoinformatik alanında protein katlanması tanıma probleminin çözümüne yeni bir yaklaşım sunularak literatüre katkı sağlanması amaçlanmıştır.
