ENSEMBLE MACHINE LEARNING TECHNIQUES FOR GENOMIC VARIANT PATHOGENICITY PREDICTION

Andrii Smoliak, Ivan Kuno

Abstract


Background. The rapid development of next-generation sequencing (NGS) technologies has resulted in a massive accumulation of genomic data, creating new opportunities and challenges for identifying genetic variants associated with diseases. Traditional statistical and rule-based approaches often fail to capture the complex, non-linear relationships between genomic features and variant pathogenicity. The integration of ensemble machine learning techniques provides a promising pathway to enhance prediction accuracy and interpretability in genomic research. This study focuses on developing an ensemble-based framework for the classification of genetic variants using gradient boosting.

Materials and Methods. The dataset was obtained from the open-access ClinVar repository, containing annotated genomic variants with clinical significance labels. After preprocessing and feature engineering, a set of ensemble algorithms, including HistGradientBoosting, XGBoost, and LightGBM, was trained and evaluated. The implemented pipeline was designed to be reproducible and computationally efficient, enabling model training on standard workstation environments.

Results and Discussion. The conducted experiments demonstrated the effectiveness of the proposed framework for the classification of pathogenic and benign genetic variants. The results indicate that the integration of biologically informed genomic features contributes to improved predictive accuracy, robustness, and generalization across heterogeneous genomic datasets. The combination of an expanded feature space with gradient boosting ensemble algorithms achieved strong ROC-AUC and PR-AUC performance on an independent test.

Conclusion. Ensemble learning represents an effective and interpretable approach for genomic variant classification and pathogenicity prediction. The study demonstrated that the integration of biologically informed feature engineering with gradient boosting ensemble models improves predictive robustness, generalization, and resilience to class imbalance in heterogeneous genomic datasets. Future work will focus on advanced deep learning components and further optimization strategies to enhance predictive performance on large-scale genomic datasets.

Keywords: ensemble learning, genomic prediction, pathogenic variants, gradient boosting, feature engineering, bioinformatics.


Full Text:

PDF

References


[1] The 1000 Genomes Project Consortium. (2015). A global reference for human genetic variation. Nature, 526(7571), 68–74. https://doi.org/10.1038/nature15393

[2] Libbrecht, M. W., & Noble, W. S. (2015). Machine learning applications in genetics and genomics. Nature Reviews Genetics, 16(6), 321–332. https://doi.org/10.1038/nrg3920

[3] Kircher, M., Witten, D. M., Jain, P., O’Roak, B. J., Cooper, G. M., & Shendure, J. (2014). A general framework for estimating the relative pathogenicity of human genetic variants. Nature Genetics, 46, 310–315.
https://doi.org/10.1038/ng.2892

[4] Friedman, J. H. (2001). Greedy function approximation: A gradient boosting machine. Annals of Statistics, 29(5), 1189–1232.
https://doi.org/10.1214/aos/1013203451

[5] Jain, A., et al. (2021). Classification of genetic variants using machine learning. arXiv preprint arXiv:2112.05154.
https://arxiv.org/abs/2112.05154

[6] Chawla, N. V., Bowyer, K. W., Hall, L. O., & Kegelmeyer, W. P. (2002). SMOTE: Synthetic minority over-sampling technique. Journal of Artificial Intelligence Research, 16, 321–357.
https://doi.org/10.1613/jair.953

[7] Quang, D., & Xie, X. (2015). DANN: A deep learning approach for annotating the pathogenicity of genetic variants. Genome Research, 25(10), 1554–1563. https://doi.org/10.1101/gr.184473.114

[8] Avsec, Ž., Agarwal, V., Visentin, D., Ledsam, J. R., Grabska-Barwinska, A., Taylor, K. R., Assael, Y., Jumper, J., Kohli, P., & Kelley, D. R. (2021). Effective gene expression prediction from sequence by integrating long-range interactions. Nature Methods, 18, 1196–1207.
https://doi.org/10.1038/s41592-021-01252-x

[9] Kelley, D. R., Snoek, J., & Rinn, J. L. (2016). Basset: Learning the regulatory code of the accessible genome with deep convolutional neural networks. Genome Research, 26(7), 990–999.
https://doi.org/10.1101/gr.200535.115

[10] Snoek, J., Larochelle, H., & Adams, R. P. (2012). Practical Bayesian optimization of machine learning algorithms. Advances in Neural Information Processing Systems, 25.
https://doi.org/10.48550/arXiv.1206.2944

[11] Rentzsch, P., Schubach, M., Shendure, J., & Kircher, M. (2021). CADD-Splice – improving genome-wide variant effect prediction using deep learning-derived splice scores. Genome Medicine, 13, 31.
https://doi.org/10.1186/s13073-021-00835-9

[12] Ng, P. C., & Henikoff, S. (2003). SIFT: Predicting amino acid changes that affect protein function. Nucleic Acids Research, 31(13), 3812–3814. https://doi.org/10.1093/nar/gkg509

[13] Alipanahi, B., Delong, A., Weirauch, M. T., & Frey, B. J. (2015). Predicting the sequence specificities of DNA- and RNA-binding proteins by deep learning. Nature Biotechnology, 33(8), 831–838.
https://doi.org/10.1038/nbt.3300

[14] Lundberg, S. M., & Lee, S.-I. (2017). A unified approach to interpreting model predictions. Advances in Neural Information Processing Systems, 30. https://doi.org/10.48550/arXiv.1705.07874

[15] Zhou, J., & Troyanskaya, O. G. (2015). Predicting effects of noncoding variants with deep learning–based sequence model. Nature Methods, 12(10), 931–934. https://doi.org/10.1038/nmeth.3547

[16] Eraslan, G., Avsec, Ž., Gagneur, J., & Theis, F. J. (2019). Deep learning: new computational modelling techniques for genomics. Nature Reviews Genetics, 20(7), 389–403.
https://doi.org/10.1038/s41576-019-0122-6




DOI: http://dx.doi.org/10.30970/eli.34.8

Refbacks

  • There are currently no refbacks.