Metehan Unal

Background/Objectives: The right and left mandibular third molars of the same patient may be expected to have similar radiographic characteristics. This study aimed to determine whether side-specific distributions of mandibular third-molar classifications correspond to patient-level bilateral agreement and whether identical Pederson scores may conceal differences in their underlying radiographic components. Methods: This retrospective cross-sectional study included 800 mandibular third molars from 400 patients with bilateral mandibular third molars. Panoramic radiographs were evaluated separately for each side using the Winter classification, Pell–Gregory classification, and Pederson difficulty index. Bilateral agreement was assessed using raw agreement, Cohen’s kappa, and weighted kappa coefficients, while side-specific distributions were compared using marginal homogeneity tests. Results: Raw agreement was 59.75% for Winter angulation, 56.0% for the combined Pell–Gregory classification, 69.5% for the Pederson difficulty group, and 48.75% for the total Pederson score. Kappa values were 0.42, 0.44, and 0.43 for Winter angulation, the combined Pell–Gregory classification, and the Pederson difficulty group, respectively. Complete agreement across all three systems was observed in 170 patients (42.5%). The Pell–Gregory ramus relationship showed substantial agreement (κ = 0.64), whereas depth showed moderate agreement (κ = 0.41). Left-side Pederson scores were significantly higher than right-side scores (p = 0.013), although the mean difference was only 0.15 points. Conclusions: Bilateral mandibular third molars showed only moderate patient-level agreement. Similarity in category distributions at the sample level should not be interpreted as within-patient agreement, and identical Pederson scores may conceal differences in angulation and position. Each mandibular third molar should therefore be evaluated separately, particularly in cases close to classification thresholds.

The Synthetic Biology Open Language (SBOL) is a community-driven and machine-readable standard for creating and sharing biological designs. SBOL has evolved over the years, resulting in three major versions, each introducing new features. Although SBOL3 is the latest community-adopted version, it has not been fully integrated into software tools and repositories due to the lack of essential libraries to facilitate this transition. In this paper, we present the SBOL Converter, a backward-compatibility tool that can update legacy designs to the latest SBOL version. Designs can also be converted to previous versions to allow legacy tools to be used in ongoing projects. Moreover, it incorporates existing SBOL tools, providing a unified interface to orchestrate the validation of designs and checking against various compliance rules and best practices. The converter can also process other popular biological formats such as GenBank and FASTA. As a lightweight tool, it can be integrated into engineering biology design workflows to accelerate tool development and enable the transition to the latest versions as they are released, facilitating the reusability and reproducibility of biological designs.

Large language models have transformed software engineering practices. However, generated artefacts are not always developer-friendly and may partially meet complex requirements. As the need to standardise, integrate, and develop tools in engineering biology increases, novel approaches are needed to create and maintain intuitive software sustainably. Here, we present an ontology-driven approach using large language models to create user-facing software libraries for knowledge graphs. We introduce an ontology-to-language framework to systematically map domain terms and graph structures. We then demonstrate this approach by creating an ontology for the latest Synthetic Biology Open Language standard and generating the sbol-script software library, which can be used within browsers or to develop applications with native web support. This ontology-driven software engineering approach and these resources are essential for the community and to facilitate the development of sustainable software projects. The SBOL3 Ontology and the sbol-script library are available from https://github.com/SynBioDex/sbol-owl3 and https://github.com/SynBioDex/sbol-script.

Human microbiota refers to the trillions of microorganisms that inhabit our bodies and have been discovered to have a substantial impact on human health and disease. By sampling the microbiota, it is possible to generate massive quantities of data for analysis using Machine Learning algorithms. In this study, we employed several modern Machine Learning techniques to predict Inflammatory Bowel Disease using raw sequence data. The dataset was obtained from NCBI preprocessed graph representations and converted into a structured form. Seven well-known Machine Learning frameworks, including Random Forest, Support Vector Machines, Extreme Gradient Boosting, Light Gradient Boosting Machine, Gaussian Naïve Bayes, Logistic Regression, and k-Nearest Neighbor, were used. Grid Search was employed for hyperparameter optimization. The performance of the Machine Learning models was evaluated using various metrics such as accuracy, precision, fscore, kappa, and area under the receiver operating characteristic curve. Additionally, Mc Nemar’s test was conducted to assess the statistical significance of the experiment. The data was constructed using k-mer lengths of 3, 4 and 5. The Light Gradient Boosting Machine model overperformed over other models with 67.24%, 74.63% and 76.47% accuracy for k-mer lengths of 3, 4 and 5, respectively. The LightGBM model also demonstrated the best performance in each metric. The study showed promising results predicting disease from raw sequence data. Finally, Mc Nemar’s test results found statistically significant differences between different Machine Learning approaches.

In the study, the implementation of 3D reconstruction of buildings using drones is explained. In this project, Airsim was used as the simulation environment and images were obtained from the simulation environment using OpenCV and the Meshroom software was run on these images and modeling was done in the computer environment. For real-world studies, the engineering faculty in Ankara University 50. Yıl Campus was modeled using photogrammetry technique. In the last part, the results of different modelling algorithms were compared.

Data from omics studies have been used for prediction and classification of various diseases in biomedical and bioinformatics research. In recent years, Machine Learning (ML) algorithms have been used in many different fields related to healthcare systems, especially for disease prediction and classification tasks. Integration of molecular omics data with ML algorithms has offered a great opportunity to evaluate clinical data. RNA sequence (RNA-seq) analysis has been emerged as the gold standard for transcriptomics analysis. Currently, it is being used widely in clinical research. In our present work, RNA-seq data of extracellular vesicles (EV) from healthy and colon cancer patients are analyzed. Our aim is to develop models for prediction and classification of colon cancer stages. Five different canonical ML and Deep Learning (DL) classifiers are used to predict colon cancer of an individual with processed RNA-seq data. The classes of data are formed on the basis of both colon cancer stages and cancer presence (healthy or cancer). The canonical ML classifiers, which are k-Nearest Neighbor (kNN), Logistic Model Tree (LMT), Random Tree (RT), Random Committee (RC), and Random Forest (RF), are tested with both forms of the data. In addition, to compare the performance with canonical ML models, One-Dimensional Convolutional Neural Network (1-D CNN), Long Short-Term Memory (LSTM), and Bidirectional LSTM (BiLSTM) DL models are utilized. Hyper-parameter optimizations of DL models are constructed by using genetic meta-heuristic optimization algorithm (GA). The best accuracy in cancer prediction is obtained with RC, LMT, and RF canonical ML algorithms as 97.33%. However, RT and kNN show 95.33% performance. The best accuracy in cancer stage classification is achieved with RF as 97.33%. This result is followed by LMT, RC, kNN, and RT with 96.33%, 96%, 94.66%, and 94%, respectively. According to the results of the experiments with DL algorithms, the best accuracy in cancer prediction is obtained with 1-D CNN as 97.67%. BiLSTM and LSTM show 94.33% and 93.67% performance, respectively. In classification of the cancer stages, the best accuracy is achieved with BiLSTM as 98%. 1-D CNN and LSTM show 97% and 94.33% performance, respectively. The results reveal that both canonical ML and DL models may outperform each other for different numbers of features.

Artificial Intelligence has emerged as a breakthrough in many fields including medicine and dentistry where new approaches can be employed to solve challenging decision making processes faced in the dental field. Artificial intelligence can be used as a decision support mechanism to solve the increasing population and consequently the increasing dental treatment needs. It also assists dentists in diagnosis and treatment planning stages that require expert opinion. This mini-review covers some of the recent studies in this area and envisions future directions on the use of machine learning in dental problems.

The intelligent machines concept is born in sci-fi scenarios. Today it seems to be we are much closer to realizing this idea than ever before. By imitating the human nervous system, machines can learn many things. This paper explains modern learning techniques like artificial neural networks, transfer learning. Later purposes an experiment to classify plant seedling images to test the transfer learning with two different CNN architectures. Although the architects were not actually created for this task, result were quite accurate for a different classification task.