Drug-resistant diagnosis for tuberculosis using DNA sequencing and machine learning Yang Yang, Katherine E Niehaus and David A Clifton

7th, July 2016

1

Yang Yang  Research interests Signal processing, nonlinear system identification, feature selection, condition monitoring

 Education 2006.09-2013.06 School of mechanical engineering, Shanghai Jiao Tong University

PhD

2002.09-2006.07 School of mechanical engineering, Shanghai Jiao Tong University

Bachelor

 Experience 2015.08-2017.08 K.C. Wang Fellowship

Computational Health Informatics laboratory, Institute of Biomedical Engineering, University of Oxford, UK 2013.06-2015.06 Post doctor School of mechanical engineering, Shanghai Jiao Tong University 2007.10-2008.10 Exchange plan 2

Center of Intelligent Maintenance System, University of Cincinnati, USA

PUBLICATION 1. Yang, Y., Z. K. Peng, X. J. Dong, W. M. Zhang, and G. Meng. "Nonlinear time-varying vibration system identification 2. 3. 4.

5. 6. 7.

using parametric time–frequency transform with spline kernel." Nonlinear Dynamics , pp. 1-16 2016. Yang Y, Peng ZK, Zhang WM, Meng G, Lang ZQ. Dispersion analysis for broadband guided wave using generalized warblet transform. Journal of Sound and Vibration. 367, pp. 22-36 2016. Yang Y., Dong X.J., Peng Z.K., Zhang W. M., Meng G., Vibration signal analysis using parameterized time-frequency method for feature extraction of varying-speed rotary machinery, Journal of Sound and Vibration，332(20), pp 350-366, 2015. Yang Y., Dong X.J., Zhang W.M., Peng Z.K., Meng G., Component Extraction for Non-stationary Multi-component Signal Using Parameterized De-chirping and Band-pass Filter, IEEE Signal Processing Letters, 2015. Yang Y., Peng Z.K., Dong X.J., Zhang W.M., General parameterized time-frequency transform, IEEE Transactions on Signal Processing, 62(11), pp 2751-2764, 2014. Yang Y., Peng Z.K., Dong X.J., Zhang W.M., Application of parameterized time-frequency analysis on multicomponent frequency modulated signals, IEEE Transactions on Instrumentation and Measurement, 63(12), pp 3169-3180, 2014. Yang Y., Zhang W.M., Peng Z.K., Meng G., Multicomponent signal analysis based on polynomial chirplet transform, IEEE Transactions on Industrial Electronics, , 60(9), pp 3948-3956, 2013.

PUBLICATION 8. Yang Y., Peng Z.K., Zhang W.M., Meng G., Spline-kernelled chirplet transform for the analysis of signals with timevarying frequency and its application, IEEE Transactions on Industrial Electronics, 59(3), pp 1612-1621, 2012. 9. Yang Y., Peng Z.K., Zhang W.M., Meng G., Frequency-varying group delay estimation using frequency domain polynomial chirplet transform, Mechanical Systems and Signal Processing, 46(1), pp 146-162, 2014. 10. Yang Y., Peng Z.K., Zhang W.M., Meng G., Characterize highly oscillating frequency modulation using generalized Warblet transform, Mechanical Systems and Signal Processing, 26, pp 128-140, 2012. 11. Yang Y., Liao Y.X., Meng G., Zhang W.M., A hybrid feature selection scheme for unsupervised learning and its application in bearing fault diagnosis, Expert Systems with Applications, 38(9), pp 11311-11320, 2011. 12. Chen, S., Yang, Y., Wei, K., Dong, X., Peng, Z. and Zhang, W., time-varying frequency-modulated component extraction based on parameterized demodulation and singular value decomposition. IEEE Transactions on Instrumentation and Measurement, 65(2), pp 276 - 285, 2016. 13. Deng, Y., Cheng, C. M., Yang, Y., Peng, Z. K., Yang, W. X., & Zhang, W. M. Parametric identification of Nonlinear Vibration Systems via Polynomial Chirplet Transform. Journal of Vibration and Acoustics. (2016). 14. Chen, S., Peng Z.K., Yang, Y., Dong, X., and Zhang, W., Chirplet Path Fusion for the Analysis of Time-Varying Frequency Modulated Signals, , IEEE Transactions on Industrial Electronics, Accepted.

Purpose

•

Resistance prediction using machine learning classifiers of multivariate association

• •

Prediction comparison varying feature sets Prediction comparison for subclades

5

Data description Frozen sputum samples from 1839 patients diagnosed with active TB from Birmingham, Oxford and surrounding regions in the UK.

All isolates are categorized into total 9 clades. •

Beijng, Delhi_CAS, EAI, EuroAmer, LAM, Tur and Uganda

Feature space is formed by total 2629 SNPs in 23 candidate genes. •

ahpC, eis, embA, embB, embC, embR, fabG1, gidB, gyrA, gyrB, inhA, iniC, katG, manB, ndh, pncA, rmlD, rpoB, rpsA, rpsL, rrs, tlyA

6

Phenotype Up to 11 drugs were assayed INH, RIF, EMB, PZA, CIP, MOX, OFX, SM, AK, CAP, KAN

7

Analysis methods Sparse logistic PCA Direct association

Classifiers •

Product of Marginal (PM)

•

Class-conditional Bernoulli mixture model (CBMM)

8

Principle component analysis

•

Suppose data y1,… yn are the n data points and considered a k- dimentional (k5%

increase