Mert Gençtürk

Copy number variants (CNVs) are established contributors to rare genetic disorders, yet their clinical interpretation remains challenging in diagnostic genomics. Large CNVs frequently encompass multiple functional regions whose clinical significance can only be resolved in the context of the patients phenotype. Effective prioritization demands variant-level scoring of dosage sensitivity, structural consequences, and disease associations, and systematic comparison of candidates within the same clinical context. Current computational tools only partially address these requirements: they automate variant-level scoring but leave phenotype-guided evidence integration and cross-variant ranking to the clinician, creating a gap between annotation throughput and diagnostic decision-making. Agentic AI systems coordinate large language model-driven reasoning across structured multi-step pipelines and have shown strong performance on biomedical tasks requiring iterative evidence evaluation and contextual judgement, making them well suited to patient-specific variant interpretation where rigid scoring functions fall short. Here, we present CoNVict, a two-stage agentic AI system for patient-specific CNV prioritization. The system ranks CNVs through verdict classification that triages candidates and tournament ranking that performs pairwise comparisons via structured, in-context reasoning. Evaluated on simulated diagnostic cases spanning multiple clinical subspecialties, CoNVict substantially outperforms existing computational methods in identifying the causal CNV and maintains robust performance on variants of uncertain significance and non-coding variants without retraining. Our results demonstrate that agentic AI can bridge the gap between automated variant-level annotation and the patient-specific clinical reasoning required for CNV-driven genetic diagnosis.

The quantification of the precise copy number variations (CNVs) is crucial to understanding the effects of gene dosage, disease severity, and therapeutic response. Although whole-exome sequencing (WES) offers a cost-effective solution for CNV detection in a clinical setting, it introduces several biases, including those related to sequence length, GC content, and the use of targeting probes. Consequently, estimating exact copy numbers remains challenging, especially for WES data. Here, we present ExactCN, a deep learning–based method for estimation of exact copy numbers from WES data per exon. The architecture integrates convolutional layers that extract local read-depth patterns with transformer encoder blocks that capture genomic context and handle sequencing noise. ExactCN is trained on WES samples from the 1000 Genomes Project, using matching WGS-based calls as semi–ground truth. In benchmarks, ExactCN improves the state-of-the-art integer CNV calling performance by reducing the macro-averaged mean absolute error (MAE) from 0.91 to 0.62 and the macro-averaged root mean squared error (RMSE) from 1.31 to 0.78. It also achieves an overall Pearson correlation of 0.669 and Spearman correlation of 0.550, improving the second-best method by 0.641 and 0.482, respectively. Furthermore, a fine-tuned and specialized version of ExactCN for aggregate CNV detection in clinically important duplicated genes SMN1/2 achieved a macro averaged F1-score of 0.657, and mean absolute error of 0.3. These results substantially improves the state-of-the-art performance and demonstrates the model's applicability to both research and clinical genomic analyses.

Data normalization is a crucial preprocessing step for enhancing model performance and training stability. In federated learning (FL), where data remains distributed across multiple parties during collaborative model training, normalization presents unique challenges due to the decentralized and often heterogeneous nature of the data. Traditional methods rely on either independent client-side processing, i.e., local normalization, or normalizing the entire dataset before distributing it to parties, i.e., pooled normalization. Local normalization can be problematic when data distributions across parties are non-IID, while the pooled normalization approach conflicts with the decentralized nature of FL. In this paper, we explore the adaptation of widely used normalization techniques to FL and define the term federated normalization. Federated normalization simulates pooled normalization by enabling the collaborative exchange of normalization parameters among parties. Thus, it achieves performance on par with pooled normalization without compromising data locality. However, sharing normalization parameters such as the mean introduces potential privacy risks, which we further mitigate through a robust privacy-preserving solution. Our contributions include: (i) We systematically evaluate the impact of various federated and local normalization techniques in heterogeneous FL scenarios, (ii) We propose a novel homomorphically encrypted k-th ranked element (and median) calculation tailored for the federated setting, enabling secure and efficient federated normalization, (iii) We propose privacy-preserving implementations of widely used normalization techniques for FL, leveraging multiparty fully homomorphic encryption (MHE).