http://repository.iiitd.edu.in/xmlui/handle/123456789/1243 CB Sun, 20 Sep 2026 13:32:14 GMT 2026-09-20T13:32:14Z http://repository.iiitd.edu.in/xmlui/handle/123456789/2140 Evaluating the quality of ontologies Sammi, Daksh; Bhushan, Lakshay; Mutharaju, Raghava (Advisor); Shimizu, Cogan (Advisor) Ontologies serve as fundamental tools for organizing and representing knowledge in various do mains. Evaluating ontology quality is crucial for ensuring their e!ectiveness and usability. This paper presents a comprehensive review of ontology evaluation framework, encompassing both traditional approaches and emerging techniques. Method-based evaluation, focusing on criteria such as syntax, structure, and semantics, o!ers insights into ontology design and construction. Additionally, level-based evaluation divides assessment into distinct layers, enhancing the un derstanding of ontology characteristics. Our tool will combine the major frameworks used by knowledge experts to evaluate their ontologies like OQuaRE, OntoQA, Orme et. al., etc., which are selected after several literature reviews about their use cases. Our tool is complemented with an additional layer of LLMs suggesting knowledge experts on how to improve their ontol ogy based on the calculated metrics and o!ering in-depth insights from each calculated metric. The LLMs in our tool are connected via a retrieval augmentation pipeline. Our tool is built to overcome the limitations of pre-exisitng tools which are either not working, not available, do not take all major frameworks into account or do not provide insights regarding calculated metrics for the knowledge expert to improve their ontology. Our paper concludes with insights into future research directions and the importance of adopting a holistic approach to ontology evaluation. Wed, 27 Nov 2024 00:00:00 GMT http://repository.iiitd.edu.in/xmlui/handle/123456789/2140 2024-11-27T00:00:00Z http://repository.iiitd.edu.in/xmlui/handle/123456789/2139 Cryo-EM resolution improvement through enhanced deep learning methods. Jain, Prem Kamal; Garg, Manan; Jain, Vivek; Kumar, Vibhor (Advisor) Recent advancements in cryo-electron microscopy (cryo-EM) have transformed the study of biomolecular structures by enabling the determination of complex structures at near-atomic res olution. However, many cryo-EM maps remain at resolutions that challenge accurate structural modeling. This project explores the application of deep learning techniques for enhancing the resolution of cryo-EM density maps, thereby improving their suitability for downstream protein structure modeling. By testing existing state-of-the-art models such as DeepEMhancer and EM-GAN, we aim to quantify improvements in structural interpretability and assess scenarios where these models underperform. Through this study, we propose to analyze and address the limitations of current models, eventually contributing to the development of more robust tools for resolution enhancement in structural biology. Wed, 27 Nov 2024 00:00:00 GMT http://repository.iiitd.edu.in/xmlui/handle/123456789/2139 2024-11-27T00:00:00Z http://repository.iiitd.edu.in/xmlui/handle/123456789/2057 Function prediction of protein cavities using 3D shape descriptors and machine learning Mittal, Pratham; Ray, Arjun (Advisor) Understanding protein function at the structural level requires accurate representation of their biologically active forms. However, structures deposited in the Protein Data Bank (PDB) often include only the asymmetric unit from crystallographic experiments, leading to discrepancies with the true oligomeric state. Such inconsistencies introduce significant challenges in down- stream analyses, particularly in cavity detection, where inter-subunit voids may be misidentified as functional binding sites. This work aims to address these challenges by constructing a curated dataset of protein struc- tures that accurately represent their functional oligomeric assemblies. PDB files were prepro- cessed to eliminate structural artifacts and to reconstruct biologically relevant assemblies. The cleaned structures were then analyzed using the CICLOP tool to identify cavity-lining residues, from which geometric and residue-level features were extracted. These features will serve as the foundation for machine learning models aimed at predicting protein function based on cavity characteristics. By integrating structural data correction with robust cavity analysis, this research establishes a reliable and interpretable pipeline for protein function prediction, with potential applications in structural bioinformatics and drug discovery. Fri, 18 Jul 2025 00:00:00 GMT http://repository.iiitd.edu.in/xmlui/handle/123456789/2057 2025-07-18T00:00:00Z http://repository.iiitd.edu.in/xmlui/handle/123456789/2056 Embedding morality in AI systems Goel, Rahul; Mutharaju, Raghava (Advisor) The increasing integration of Artificial Intelligence into complex societal domains necessitates a deeper understanding of how to embed ethical reasoning into machines. This project addresses the challenge of quantifying and predicting alignment with established human ethical frameworks. Our methodology involved three key stages. First, we established a granular theoretical framework by defining 15 distinct ethical theories under the three major schools of thought: Consequentialism (α), Deontology (β), and Virtue Ethics (γ). Second, we developed a novel dataset of 450 ethical scenarios, specifically designed to elicit responses corresponding to each of the 15 theories. These scenarios were then annotated using Large Language Models (LLMs) to generate quantitative scores (α, β, γ) representing the relevance of each ethical school to a given case. Finally, we benchmarked a suite of classical machine learning models to predict these ethical alignments from textual features. Two primary experiments were conducted: a multi-output regression task to predict the (α, β, γ) scores and a multi-output classification task to predict the specific ethical school and theory. In the regression task, Linear Regression demonstrated the strongest explanatory power, achieving an R-squared (R2 ) value of 0.5804. For classification, Logistic Regression was the top- performing model, achieving an Exact Match Accuracy of 74.44% and 100% accuracy in identifying the high-level ethical school. This work provides a foundational dataset, a robust evaluation of classical machine learning models for ethical prediction, and confirms that quantitative representations of ethical schools are viable targets for computational modeling. The findings lay the groundwork for future research in building more transparent, auditable, and ethically-aligned AI systems. Fri, 18 Jul 2025 00:00:00 GMT http://repository.iiitd.edu.in/xmlui/handle/123456789/2056 2025-07-18T00:00:00Z