MEFY Gen Varyantlarında Modifiye Edilmiş Sert Oylama Sınıflandırıcısı Uygulaması, In-Silico Araç Performansını Artırıyor: Küçük Örneklem Boyutu İçin Yeni Bir Yaklaşım

Yıl 2025, Cilt: 8 Sayı: 1, 35 - 46, 18.03.2025

Tarık Alay , İbrahim Demir , Murat Kirisci

https://doi.org/10.38016/jista.1501164

Öz

Amaç: MEFV geninde bilinen sınırlı sayıda patojenik varyant bulunmaktadır. İn siliko araçlar, birçok MEFV gen varyantını sınıflandıramamaktadır. Bu nedenle, yeni yaklaşımların uygulanması gerekmektedir. Sert oylama sınıflandırıcıları ve sağlam doğrulama teknikleri sınıflandırma için kullanılabilir; ancak çift sayı sınıflandırması doğru bir şekilde yapılamamaktadır. Amacımız, hem çift sayı sınıflandırma sorununu çözmek hem de küçük veri setleri kullanarak MEFV gen varyantı tahmin doğruluğunu artırmak için yeni bir strateji geliştirmektir.
Yöntem: İlk olarak model için optimal sayıda hesaplama aracını belirledik. Daha sonra, belirlenen araçlar kullanılarak MEFV gen varyantlarını içeren eğitim veri setinde sekiz farklı makine öğrenme algoritması uygulandı. Eğitim ve doğrulama veri setinin kullanımıyla, modifiye edilmiş sert oylama makine öğrenme algoritmalarının uygulanmasına başlandı. Bundan sonra, tahmin sonuçları ile mevcut algoritmalar ve çalışmalar arasında karşılaştırmalı bir analiz gerçekleştirildi. Son olarak, gen ve protein düzeyinde değerlendirme yapılarak hotspot bölgeler belirlendi.
Bulgular: Topluluk sınıflandırıcısı, ortalama ROC AUC puanlarının %88 olduğunu gösterdi ve modifiye edilmiş sert oylama sınıflandırıcı yöntemi ile bilinen tüm varyantları %82 doğrulukla sınıflandırdı. Bu oran, hem yumuşak (%75) hem de sert oylama sınıflandırıcı (%70) yöntemlerinden daha yüksektir. Tüm varyantların kolektif değerlendirilmesi, LP varyantlarının, LB varyantlarına göre alanlarda yaklaşık 2,5 kat daha yaygın olduğunu ortaya koymuştur (χ2:13.574, p < 0.001, OR: 2.509 [1.532-4.132]).
Sonuç: MEFV gen mutasyonlarının klinik sonuçlarıyla ilgili bilgi yetersizliği göz önüne alındığında, modifiye edilmiş sert oylama sınıflandırıcı yaklaşımını kullanmak, hesaplama araçlarının sınıflandırma doğruluğunu artırmak için küçük örneklemlerde makul bir yöntem olabilir.

Anahtar Kelimeler

Sınıflandırma, Ailevi Akdeniz Ateşi, Makine Öğrenmesi, MEFV, Oylama Sınıflandırıcısı

Proje Numarası

1

Modified Hard Voting Classifier Implementation on MEFV Gene Variants Increases in Silico Tool Performance: A Novel Approach for Small Sample Size

Öz

Objective: There are a limited number of pathogenic variants known in the MEFV gene. In silico tools fail to classify many MEFV gene variants. Therefore, it is essential to implement novel approaches. Our goal is to develop a new strategy to solve the even number classification problem while improving MEFV gene variant prediction accuracy using small datasets.
Material - methods: First, we determined the optimal number of computational tools for the model. We then applied eight distinct ML algorithms on the training dataset containing MEFV gene variants using the determined tools. We initiated the application of modified hard voting machine learning algorithms, using a training and validation dataset. Subsequently, we implemented a comparative analysis between the prediction results and existing algorithms and studies. Finally, we evaluated the gene and protein level ascertainment to identify hotspot regions.
Results: The ensemble classifier scored an average ROCAUC of 88%. The modified hard voting method correctly classified all known variants with 82% accuracy, outperforming both the soft voting (75%) and hard voting (70%) methods. The results showed that the prevalence of LP variants was approximately 2.5 times higher in domains compared to LB variants(χ2: 13.574, p < 0.001, OR: 2.509 [1.532-4.132]).
Conclusion: Considering the limited understanding of the clinical implications associated with MEFV gene mutations, employing a modified hard voting classifier approach may improve the classification accuracy of computational tools.

Anahtar Kelimeler

Classification, FMF, Machine Learning, MEFV, Voting Classifier

Etik Beyan

No ethics committee approval is needed. Open source

Destekleyen Kurum

None

Proje Numarası

1

Kaynakça

• Accetturo, M. et al. (2020) ‘Improvement of MEFV gene variants classification to aid treatment decision making in familial Mediterranean fever.’, Rheumatology (Oxford, England), 59(4), pp. 754–761. Available at: https://doi.org/10.1093/rheumatology/kez332.
• Accetturo, M., Bartolomeo, N. and Stella, A. (2020) ‘In-silico Analysis of NF1 Missense Variants in ClinVar: Translating Variant Predictions into Variant Interpretation and Classification.’, International journal of molecular sciences, 21(3). Available at: https://doi.org/10.3390/ijms21030721.
• Acharjee, A. et al. (2020) ‘A random forest based biomarker discovery and power analysis framework for diagnostics research’, BMC Medical Genomics, 13(1), p. 178. Available at: https://doi.org/10.1186/s12920-020-00826-6.
• Adzhubei, I., Jordan, D.M. and Sunyaev, S.R. (2013) ‘Predicting functional effect of human missense mutations using PolyPhen-2.’, Current protocols in human genetics, Chapter 7, p. Unit7.20. Available at: https://doi.org/10.1002/0471142905.hg0720s76.
• Alay, M.T. (2024) ‘An Ensemble Model Based on Combining BayesDel and Revel Scores Indicates Outstanding Performance: Importance of Outlier Detection and Comparison of Models’, Cerrahpasa Medical Journal, 48(2), pp. 179–184.
• Albaradei, S. et al. (2021) ‘Machine learning and deep learning methods that use omics data for metastasis prediction.’, Computational and structural biotechnology journal, 19, pp. 5008–5018. Available at: https://doi.org/10.1016/j.csbj.2021.09.001.
• Awe, O.O. et al. (2024) ‘Weighted hard and soft voting ensemble machine learning classifiers: Application to anaemia diagnosis’, in Sustainable Statistical and Data Science Methods and Practices: Reports from LISA 2020 Global Network, Ghana, 2022. Springer, pp. 351–374.
• Burdon, K.P. et al. (2022) ‘Specifications of the ACMG/AMP variant curation guidelines for myocilin: Recommendations from the clingen glaucoma expert panel.’, Human mutation, 43(12), pp. 2170–2186. Available at: https://doi.org/10.1002/humu.24482.
• Cheng, J. et al. (2023) ‘Accurate proteome-wide missense variant effect prediction with AlphaMissense.’, Science (New York, N.Y.), 381(6664), p. eadg7492. Available at: https://doi.org/10.1126/science.adg7492.
• Dalmaijer, E.S., Nord, C.L. and Astle, D.E. (2022) ‘Statistical power for cluster analysis’, BMC Bioinformatics, 23(1), pp. 1–28. Available at: https://doi.org/10.1186/s12859-022-04675-1.
• Dundar, M. et al. (2022) ‘Clinical and molecular evaluation of MEFV gene variants in the Turkish population: a study by the National Genetics Consortium.’, Functional & integrative genomics, 22(3), pp. 291–315. Available at: https://doi.org/10.1007/s10142-021-00819-3.
• El-Sofany, H., Bouallegue, B. and El-Latif, Y.M.A. (2024) ‘A proposed technique for predicting heart disease using machine learning algorithms and an explainable AI method.’,Scientific reports, 14(1), p. 23277. Available at: https://doi.org/10.1038/s41598-024-74656-2.
• Fortuno, C. et al. (2018) ‘Improved, ACMG-compliant, in silico prediction of pathogenicity for missense substitutions encoded by TP53 variants.’, Human mutation, 39(8), pp. 1061–1069. Available at: https://doi.org/10.1002/humu.23553.
• Van Gijn, M.E. et al. (2018) ‘New workflow for classification of genetic variants’ pathogenicity applied to hereditary recurrent fevers by the International Study Group for Systemic Autoinflammatory Diseases (INSAID).’, Journal of medical genetics, 55(8), pp. 530–537. Available at: https://doi.org/10.1136/jmedgenet-2017-105216.
• Grandemange, S. et al. (2011) ‘The regulation of MEFV expression and its role in health and familial Mediterranean fever’, Genes & Immunity, 12(7), pp. 497–503. Available at: https://doi.org/10.1038/gene.2011.53.
• Gunning, A.C. et al. (2021) ‘Assessing performance of pathogenicity predictors using clinically relevant variant datasets’, Journal of Medical Genetics, 58(8), pp. 547–555. Available at: https://doi.org/10.1136/jmedgenet-2020-107003.
• Harrison, S.M., Biesecker, L.G. and Rehm, H.L. (2019) ‘Overview of Specifications to the ACMG/AMP Variant Interpretation Guidelines.’, Current protocols in human genetics, 103(1), p. e93. Available at: https://doi.org/10.1002/cphg.93.
• Hu, Y.-H. et al. (2024) ‘A novel MissForest-based missing values imputation approach with recursive feature elimination in medical applications’, BMC Medical Research Methodology, 24(1), p. 269. Available at: https://doi.org/10.1186/s12874-024-02392-2.
• Ioannidis, N.M. et al. (2016) ‘REVEL: An Ensemble Method for Predicting the Pathogenicity of Rare Missense Variants.’, American journal of human genetics, 99(4), pp. 877–885. Available at: https://doi.org/10.1016/j.ajhg.2016.08.016.
• Khalid, Z. and Sezerman, O.U. (2018) ‘Computational drug repurposing to predict approved and novel drug-disease associations’, Journal of Molecular Graphics and Modelling, 85, pp. 91–96. Available at: https://doi.org/https://doi.org/10.1016/j.jmgm.2018.08.005.
• Kırnaz, B., Gezgin, Y. and Berdeli, A. (2022) ‘MEFV gene allele frequency and genotype distribution in 3230 patients’ analyses by next generation sequencing methods.’, Gene, 827, p. 146447. Available at: https://doi.org/10.1016/j.gene.2022.146447.
• Knecht, C. et al. (2017) ‘IMHOTEP-a composite score integrating popular tools for predicting the functional consequences of non-synonymous sequence variants.’, Nucleic acids research, 45(3), p. e13. Available at: https://doi.org/10.1093/nar/gkw886.
• Lai, A. et al. (2022) ‘The ClinGen Brain Malformation Variant Curation Expert Panel: Rules for somatic variants in AKT3, MTOR, PIK3CA, and PIK3R2.’, Genetics in medicine : official journal of the American College of Medical Genetics, 24(11), pp. 2240–2248. Available at: https://doi.org/10.1016/j.gim.2022.07.020.
• Larracy, R., Phinyomark, A. and Scheme, E. (2021) ‘Machine learning model validation for early stage studies with small sample sizes’, in 2021 43rd Annual International Conference of the IEEE Engineering in Medicine & Biology Society (EMBC). IEEE, pp. 2314–2319.
• Liu, C. et al. (2013) ‘Applications of machine learning in genomics and systems biology.’, Computational and mathematical methods in medicine, p. 587492. Available at: https://doi.org/10.1155/2013/587492.
• Liu, X. et al. (2016) ‘dbNSFP v3.0: A One-Stop Database of Functional Predictions and Annotations for Human Nonsynonymous and Splice-Site SNVs.’, Human mutation, 37(3), pp. 235–241. Available at: https://doi.org/10.1002/humu.22932.
• Luan, J. et al. (2020) ‘The predictive performances of random forest models with limited sample size and different species traits’, Fisheries Research, 227, p. 105534. Available at: https://doi.org/https://doi.org/10.1016/j.fishres.2020.105534.
• Megantara, A.A. and Ahmad, T. (2021) ‘A hybrid machine learning method for increasing the performance of network intrusion detection systems’, Journal of Big Data, 8(1). Available at: https://doi.org/10.1186/s40537-021-00531-w.
• Mighton, C. et al. (2022) ‘Data sharing to improve concordance in variant interpretation across laboratories: results from the Canadian Open Genetics Repository’, Journal of Medical Genetics, 59(6), pp. 571 LP – 578. Available at: https://doi.org/10.1136/jmedgenet-2021-107738.
• Ng, P.C. and Henikoff, S. (2003) ‘SIFT: Predicting amino acid changes that affect protein function.’, Nucleic acids research, 31(13), pp. 3812–3814. Available at: https://doi.org/10.1093/nar/gkg509.
• Nykamp, K. et al. (2017) ‘Sherloc: a comprehensive refinement of the ACMG-AMP variant classification criteria.’, Genetics in medicine : official journal of the American College of Medical Genetics, 19(10), pp. 1105–1117. Available at: https://doi.org/10.1038/gim.2017.37.
• Ogundimu, E.O., Altman, D.G. and Collins, G.S. (2016) ‘Adequate sample size for developing prediction models is not simply related to events per variable.’, Journal of clinical epidemiology, 76, pp. 175–182. Available at: https://doi.org/10.1016/j.jclinepi.2016.02.031.
• Palanivinayagam, A. and Damaševičius, R. (2023) ‘Effective Handling of Missing Values in Datasets for Classification Using Machine Learning Methods’, Information, 14(2), p. 92.
• Papin, S. et al. (2007) ‘The SPRY domain of Pyrin, mutated in familial Mediterranean fever patients, interacts with inflammasome components and inhibits proIL-1beta processing.’, Cell death and differentiation, 14(8), pp. 1457–1466. Available at: https://doi.org/10.1038/sj.cdd.4402142.
• Pejaver, V. et al. (2022) ‘Calibration of computational tools for missense variant pathogenicity classification and ClinGen recommendations for PP3/BP4 criteria.’, American journal of human genetics, 109(12), pp. 2163–2177. Available at: https://doi.org/10.1016/j.ajhg.2022.10.013.
• Pyeritz, R.E. and for the Professional Practice and Guidelines Committee, A. (2012) ‘Evaluation of the adolescent or adult with some features of Marfan syndrome’, Genetics in Medicine, 14(1), pp. 171–177. Available at: https://doi.org/10.1038/gim.2011.48.
• Rajput, D., Wang, W.-J. and Chen, C.-C. (2023) ‘Evaluation of a decided sample size in machine learning applications’, BMC Bioinformatics, 24(1), p. 48. Available at: https://doi.org/10.1186/s12859-023-05156-9.
• Richards, S. et al. (2015) ‘Standards and guidelines for the interpretation of sequence variants: a joint consensus recommendation of the American College of Medical Genetics and Genomics and the Association for Molecular Pathology.’, Genetics in medicine : official journal of the American College of Medical Genetics, 17(5), pp. 405–424. Available at: https://doi.org/10.1038/gim.2015.30.
• Riley, R.D. et al. (2019) ‘Minimum sample size for developing a multivariable prediction model: PART II‐binary and time‐to‐event outcomes’, Statistics in medicine, 38(7), pp. 1276–1296.
• Sallah, S.R. et al. (2022) ‘Improving the clinical interpretation of missense variants in X linked genes using structural analysis.’, Journal of medical genetics, 59(4), pp. 385–392. Available at: https://doi.org/10.1136/jmedgenet-2020-107404.
• Savige, J. et al. (2021) ‘Consensus statement on standards and guidelines for the molecular diagnostics of Alport syndrome: refining the ACMG criteria.’, European journal of human genetics : EJHG, 29(8), pp. 1186–1197. Available at: https://doi.org/10.1038/s41431-021-00858-1.
• Song, X. et al. (2021) ‘Comparison of machine learning and logistic regression models in predicting acute kidney injury: A systematic review and meta-analysis.’, International journal of medical informatics, 151, p. 104484. Available at: https://doi.org/10.1016/j.ijmedinf.2021.104484.
• Stewart, D.R. et al. (2018) ‘Care of adults with neurofibromatosis type 1: a clinical practice resource of the American College of Medical Genetics and Genomics (ACMG)’, Genetics in Medicine, 20(7), pp. 671–682. Available at: https://doi.org/10.1038/gim.2018.28.
• Tian, Y. et al. (2019) ‘REVEL and BayesDel outperform other in silico meta-predictors for clinical variant classification’, Scientific Reports, 9(1), p. 12752. Available at: https://doi.org/10.1038/s41598-019-49224-8.
• Vabalas, A. et al. (2019) ‘Machine learning algorithm validation with a limited sample size.’, PloS one, 14(11), p. e0224365. Available at: https://doi.org/10.1371/journal.pone.0224365.
• Vu, T.T. and Braga-Neto, U.M. (2009) ‘Is bagging effective in the classification of small-sample genomic and proteomic data?’, EURASIP journal on bioinformatics & systems biology, 2009(1), p. 158368. Available at: https://doi.org/10.1155/2009/158368.
• Waring, A. et al. (2021) ‘Data-driven modelling of mutational hotspots and in silico predictors in hypertrophic cardiomyopathy.’, Journal of medical genetics, 58(8), pp. 556–564. Available at: https://doi.org/10.1136/jmedgenet-2020-106922.
• Wilcox, E.H. et al. (2022) ‘Evaluating the impact of in silico predictors on clinical variant classification.’, Genetics in medicine : official journal of the American College of Medical Genetics, 24(4), pp. 924–930. Available at: https://doi.org/10.1016/j.gim.2021.11.018.