A Machine Learning Approach for the Classification of Kidney Cancer Subtypes Using miRNA Genome Data
Abstract
:1. Introduction
2. The RNA Sequence and Kidney Cancer
3. Machine Learning
3.1. Neighborhood Component Analysis
3.2. LSTM
- is the activation vector of the forget gate,
- is the sigmoid function,
- W is weight matrices to be learned during training,
- is input vector to the LSTM unit,
- b is bias vector parameters to be learned during training,
- is activation vector of the input gate,
- is cell state vector,
- is activation vector of the output gate, and
- is output vector of the LSTM unit.
4. Data Preparation and Results
4.1. Data Preparation and Categorization
4.2. Results and Discussions
5. Conclusions
Author Contributions
Funding
Acknowledgments
Conflicts of Interest
References
- Verbiest, A.; Couchy, G.; Job, S.; Caruana, L.; Lerut, E.; Oyen, R.; de Reyniès, A.; Tosco, L.; Joniau, S.; Van Poppel, H.; et al. Molecular subtypes of clear-cell renal cell carcinoma are prognostic for outcome after complete metastasectomy. Eur. Urol. 2018, 74, 474–480. [Google Scholar] [CrossRef] [PubMed]
- NCI. The NHGRI. The Cancer Genome Atlas Homepage. Available online: https://cancergenome.nih.gov/ (accessed on 28 April 2009).
- Yang, W.; Wang, K.; Zuo, W. Neighborhood Component Feature Selection for High-Dimensional Data. JCP 2012, 7, 161–168. [Google Scholar] [CrossRef]
- Hochreiter, S.; Schmidhuber, J. Long short-term memory. Neural Comput. 1997, 9, 1735–1780. [Google Scholar] [CrossRef] [PubMed]
- Ibrahim, A.K.; Zhuang, H.; Chérubin, L.M.; Schärer-Umpierre, M.T.; Erdol, N. Automatic classification of grouper species by their sounds using deep neural networks. J. Acoust. Soc. Am. 2018, 144, EL196–EL202. [Google Scholar] [CrossRef] [PubMed]
- Matthews, B.W. Comparison of the predicted and observed secondary structure of T4 phage lysozyme. Biochim. Biophys. Acta 1975, 405, 442–451. [Google Scholar] [CrossRef]
- Friedman, R.C.; Farh, K.K.H.; Burge, C.B.; Bartel, D.P. Most mammalian mRNAs are conserved targets of microRNAs. Genome Res. 2008, 19, 92–105. [Google Scholar] [CrossRef] [PubMed] [Green Version]
- Ambros, V. The functions of animal microRNAs. Nature 2004, 431, 350. [Google Scholar] [CrossRef] [PubMed]
- Jansson, M.D.; Lund, A.H. MicroRNA and cancer. Mol. Oncol. 2012, 6, 590–610. [Google Scholar] [CrossRef] [PubMed] [Green Version]
- Croce, C.M. Causes and consequences of microRNA dysregulation in cancer. Nat. Rev. Genet. 2009, 10, 704. [Google Scholar] [CrossRef] [PubMed]
- Lu, J.; Getz, G.; Miska, E.A.; Alvarez-Saavedra, E.; Lamb, J.; Peck, D.; Sweet-Cordero, A.; Ebert, B.L.; Mak, R.H.; Ferrando, A.A.; et al. MicroRNA expression profiles classify human cancers. Nature 2005, 435, 834. [Google Scholar] [CrossRef] [PubMed]
- Munker, R.; Calin, G.A. MicroRNA profiling in cancer. Clin. Sci. 2011, 121, 141–158. [Google Scholar] [CrossRef] [PubMed] [Green Version]
- Volinia, S.; Calin, G.A.; Liu, C.G.; Ambs, S.; Cimmino, A.; Petrocca, F.; Visone, R.; Iorio, M.; Roldo, C.; Ferracin, M.; et al. A microRNA expression signature of human solid tumors defines cancer gene targets. Proc. Natl. Acad. Sci. USA 2006, 103, 2257–2261. [Google Scholar] [CrossRef] [PubMed] [Green Version]
- Papadopoulos, E.I.; Petraki, C.; Gregorakis, A.; Fragoulis, E.G.; Scorilas, A. Clinical evaluation of microRNA-145 expression in renal cell carcinoma: A promising molecular marker for discriminating and staging the clear cell histological subtype. Biol. Chem. 2016, 397, 529–539. [Google Scholar] [CrossRef] [PubMed]
- White, N.M.; Bao, T.T.; Grigull, J.; Youssef, Y.M.; Girgis, A.; Diamandis, M.; Fatoohi, E.; Metias, M.; Honey, R.J.; Stewart, R.; et al. miRNA profiling for clear cell renal cell carcinoma: Biomarker discovery and identification of potential controls and consequences of miRNA dysregulation. J. Urol. 2011, 186, 1077–1083. [Google Scholar] [CrossRef] [PubMed]
- Juan, D.; Alexe, G.; Antes, T.; Liu, H.; Madabhushi, A.; Delisi, C.; Ganesan, S.; Bhanot, G.; Liou, L.S. Identification of a microRNA panel for clear-cell kidney cancer. Urology 2010, 75, 835–841. [Google Scholar] [CrossRef] [PubMed]
- Samaan, S.; Khella, H.W.; Girgis, A.; Scorilas, A.; Lianidou, E.; Gabril, M.; Krylov, S.N.; Jewett, M.; Bjarnason, G.A.; El-said, H.; et al. miR-210 is a prognostic marker in clear cell renal cell carcinoma. J. Mol. Diagn. 2015, 17, 136–144. [Google Scholar] [CrossRef] [PubMed]
- Zhang, W.; Ni, M.; Su, Y.; Wang, H.; Zhu, S.; Zhao, A.; Li, G. MicroRNAs in serum exosomes as potential biomarkers in clear-cell renal cell carcinoma. Eur. Urol. Focus 2016, 4, 412–419. [Google Scholar] [CrossRef] [PubMed]
- Vergho, D.; Kneitz, S.; Rosenwald, A.; Scherer, C.; Spahn, M.; Burger, M.; Riedmiller, H.; Kneitz, B. Combination of expression levels of miR-21 and miR-126 is associated with cancer-specific survival in clear-cell renal cell carcinoma. BMC Cancer 2014, 14, 25. [Google Scholar] [CrossRef] [PubMed]
- Zaman, M.S.; Shahryari, V.; Deng, G.; Thamminana, S.; Saini, S.; Majid, S.; Chang, I.; Hirata, H.; Ueno, K.; Yamamura, S.; et al. Correction: Up-Regulation of MicroRNA-21 Correlates with Lower Kidney Cancer Survival. PLoS ONE 2012, 7, e31060. [Google Scholar] [CrossRef]
- Wach, S.; Nolte, E.; Theil, A.; Stöhr, C.; Rau, T.; Hartmann, A.; Ekici, A.; Keck, B.; Taubert, H.; Wullich, B. MicroRNA profiles classify papillary renal cell carcinoma subtypes. Br. J. Cancer 2013, 109, 714. [Google Scholar] [CrossRef] [PubMed]
- White, N.; Khella, H.; Grigull, J.; Adzovic, S.; Youssef, Y.; Honey, R.; Stewart, R.; Pace, K.; Bjarnason, G.; Jewett, M.; et al. miRNA profiling in metastatic renal cell carcinoma reveals a tumour-suppressor effect for miR-215. Br. J. Cancer 2011, 105, 1741. [Google Scholar] [CrossRef] [PubMed]
- Youssef, Y.M.; White, N.M.; Grigull, J.; Krizova, A.; Samy, C.; Mejia-Guerrero, S.; Evans, A.; Yousef, G.M. Accurate molecular classification of kidney cancer subtypes using microRNA signature. Eur. Urol. 2011, 59, 721–730. [Google Scholar] [CrossRef] [PubMed]
- Petillo, D.; Kort, E.J.; Anema, J.; Furge, K.A.; Yang, X.J.; Teh, B.T. MicroRNA profiling of human kidney cancer subtypes. Int. J. Oncol. 2009, 35, 109–114. [Google Scholar] [CrossRef] [PubMed]
- Wang, J.; Lee, A.; Huang, M.; Ibrahim, A.K.; Zhuang, H.; Muhamed Ali, A. Classification of White Blood Cells with PatternNet-fused Ensemble of Convolutional Neural Networks (PECNN). In Proceedings of the International Symposium on Signal Processing and Information Technology (ISSPIT), Louisville, KY, USA, 6–8 December 2018. [Google Scholar]
- Wang, J.; Ibrahim, A.K.; Zhuang, H.; Muhamed Ali, A.; Li, A. A Study on Automatic Detection of IDC Breast Cancer with Convolutional Neural Networks. In Proceedings of the 2018 International Conference on Computational Science and Computational Intelligence (CSCI’18), Las Vegas, NV, USA, 13–15 December 2018. [Google Scholar]
- Mobiny, A.; Moulik, S.; Gurcan, I.; Shah, T.; Van Nguyen, H. Lung Cancer Screening Using Adaptive Memory-Augmented Recurrent Networks. arXiv, 2017; arXiv:1710.05719. [Google Scholar]
- Ypsilantis, P.P.; Montana, G. Recurrent convolutional networks for pulmonary nodule detection in CT imaging. arXiv, 2016; arXiv:1609.09143. [Google Scholar]
- Bychkov, D.; Linder, N.; Turkki, R.; Nordling, S.; Kovanen, P.E.; Verrill, C.; Walliander, M.; Lundin, M.; Haglund, C.; Lundin, J. Deep learning based tissue analysis predicts outcome in colorectal cancer. Sci. Rep. 2018, 8, 3395. [Google Scholar] [CrossRef] [PubMed]
- Zheng, Y.; Liu, D.; Georgescu, B.; Xu, D.; Comaniciu, D. Deep Learning Based Automatic Segmentation of Pathological Kidney in CT: Local Versus Global Image Context. In Deep Learning and Convolutional Neural Networks for Medical Image Computing; Springer: Berlin, Germany, 2017; pp. 241–255. [Google Scholar]
- Maggio, V.; Chierici, M.; Jurman, G.; Furlanello, C. A multiobjective deep learning approach for predictive classification in Neuroblastoma. arXiv, 2017; arXiv:1711.08198. [Google Scholar]
- DeVries, T.; Taylor, G.W. Dataset augmentation in feature space. arXiv, 2017; arXiv:1702.05538. [Google Scholar]
- Gorodkin, J. Comparing two K-category assignments by a K-category correlation coefficient. Comput. Biol. Chem. 2004, 28, 367–374. [Google Scholar] [CrossRef] [PubMed]
- Boughorbel, S.; Jarray, F.; El-Anbari, M. Optimal classifier for imbalanced data using Matthews Correlation Coefficient metric. PLoS ONE 2017, 12, e0177678. [Google Scholar] [CrossRef] [PubMed]
Disease Type | Project Name | No. of Cases | No. of Files |
---|---|---|---|
Kidney Renal Clear Cell Carcinoma | TCGA-KIRC | 516 | 616 |
Kidney Renal Papillary Cell Carcinoma | TCGA-KIRP | 291 | 323 |
High-Risk Wilms Tumor | TARGET-WT | 127 | 138 |
Kidney Chromophobe | TCGA-KICH | 66 | 91 |
Rhabdoid Tumor | TARGET-RT | 44 | 50 |
miRNA Index | miRNA Name | Feature Weight |
---|---|---|
4 | ’hsa-let-7a-1’ | 1.1857 |
10 | ’hsa-let-7a-2’ | 1.0278 |
13 | ’hsa-let-7a-3’ | 0.9616 |
14 | ’hsa-let-7b’ | 0.8815 |
15 | ’hsa-let-7c’ | 0.7158 |
16 | ’hsa-let-7d’ | 0.6871 |
23 | ’hsa-let-7e’ | 0.6759 |
24 | ’hsa-let-7f-1’ | 0.6581 |
28 | ’hsa-let-7f-2’ | 0.6572 |
78 | ’hsa-let-7g’ | 0.6518 |
159 | ’hsa-let-7i’ | 0.6412 |
161 | ’hsa-mir-100’ | 0.603 |
187 | ’hsa-mir-101-1’ | 0.5633 |
199 | ’hsa-mir-101-2’ | 0.5243 |
231 | ’hsa-mir-103a-1’ | 0.5108 |
236 | ’hsa-mir-103a-2’ | 0.5085 |
241 | ’hsa-mir-103b-1’ | 0.4886 |
248 | ’hsa-mir-103b-2’ | 0.4488 |
249 | ’hsa-mir-105-1’ | 0.4212 |
268 | ’hsa-mir-105-2’ | 0.4062 |
269 | ’hsa-mir-106a’ | 0.3732 |
270 | ’hsa-mir-106b’ | 0.3699 |
289 | ’hsa-mir-107’ | 0.3506 |
290 | ’hsa-mir-10a’ | 0.3103 |
301 | ’hsa-mir-10b’ | 0.2591 |
314 | ’hsa-mir-1-1’ | 0.2375 |
316 | ’hsa-mir-1178’ | 0.227 |
317 | ’hsa-mir-1179’ | 0.2214 |
318 | ’hsa-mir-1180’ | 0.2145 |
442 | ’hsa-mir-1181’ | 0.1944 |
453 | ’hsa-mir-1182’ | 0.1457 |
455 | ’hsa-mir-1183’ | 0.0946 |
1055 | ’hsa-mir-1184-1’ | 0.0507 |
1259 | ’hsa-mir-1184-2’ | 0.0468 |
1308 | ’hsa-mir-1184-3’ | 0.0326 |
Selected 35 miRNA for Unbalanced Classes | All 1627 miRNA for Unbalanced Classes | |||||||
---|---|---|---|---|---|---|---|---|
Class | TP | FP | FN | TN | TP | FP | FN | TN |
WT | 1331 | 40 | 48 | 10,781 | 1311 | 70 | 69 | 10,750 |
KICH | 818 | 112 | 91 | 11,179 | 769 | 174 | 139 | 11,118 |
KIRC | 5992 | 140 | 165 | 5903 | 5824 | 260 | 334 | 5782 |
KIRP | 3068 | 239 | 187 | 8706 | 2985 | 320 | 270 | 8625 |
RT | 440 | 20 | 60 | 11,680 | 444 | 43 | 55 | 11,658 |
Selected 35 miRNA for Unbalanced Classes | All 1627 miRNA for Unbalanced Classes | |||||||
---|---|---|---|---|---|---|---|---|
Class | TP | FP | FN | TN | TP | FP | FN | TN |
WT | 1309 | 41 | 71 | 10,789 | 1340 | 38 | 40 | 10,792 |
KICH | 896 | 178 | 14 | 11,122 | 877 | 74 | 33 | 11,226 |
KIRC | 5920 | 68 | 240 | 5982 | 6015 | 71 | 145 | 5979 |
KIRP | 3044 | 179 | 216 | 8771 | 3169 | 113 | 91 | 8837 |
RT | 477 | 98 | 23 | 11,612 | 467 | 46 | 33 | 11,664 |
Class | Selected 35 miRNA for Balanced Classes | 1627 miRNA for Balanced Classes | Selected 35 miRNA for Unbalanced Classes | 1627 miRNA for Unbalanced Classes |
---|---|---|---|---|
WT | 0.953 | 0.968 | 0.963 | 0.943 |
KICH | 0.898 | 0.938 | 0.880 | 0.817 |
KIRC | 0.949 | 0.964 | 0.950 | 0.902 |
KIRP | 0.917 | 0.957 | 0.911 | 0.877 |
RT | 0.884 | 0.918 | 0.914 | 0.8965 |
Over all | 0.920 | 0.949 | 0.924 | 0.887 |
© 2018 by the authors. Licensee MDPI, Basel, Switzerland. This article is an open access article distributed under the terms and conditions of the Creative Commons Attribution (CC BY) license (http://creativecommons.org/licenses/by/4.0/).
Share and Cite
Muhamed Ali, A.; Zhuang, H.; Ibrahim, A.; Rehman, O.; Huang, M.; Wu, A. A Machine Learning Approach for the Classification of Kidney Cancer Subtypes Using miRNA Genome Data. Appl. Sci. 2018, 8, 2422. https://doi.org/10.3390/app8122422
Muhamed Ali A, Zhuang H, Ibrahim A, Rehman O, Huang M, Wu A. A Machine Learning Approach for the Classification of Kidney Cancer Subtypes Using miRNA Genome Data. Applied Sciences. 2018; 8(12):2422. https://doi.org/10.3390/app8122422
Chicago/Turabian StyleMuhamed Ali, Ali, Hanqi Zhuang, Ali Ibrahim, Oneeb Rehman, Michelle Huang, and Andrew Wu. 2018. "A Machine Learning Approach for the Classification of Kidney Cancer Subtypes Using miRNA Genome Data" Applied Sciences 8, no. 12: 2422. https://doi.org/10.3390/app8122422
APA StyleMuhamed Ali, A., Zhuang, H., Ibrahim, A., Rehman, O., Huang, M., & Wu, A. (2018). A Machine Learning Approach for the Classification of Kidney Cancer Subtypes Using miRNA Genome Data. Applied Sciences, 8(12), 2422. https://doi.org/10.3390/app8122422