臺灣博碩士論文加值系統

抗萬古黴素腸球菌是一種本身對萬古黴素有抗藥性的細菌，期引起感染包括泌尿道感染、菌血症、心內膜炎等，同時也普遍在人體和自然中出現。在1985年被發現其細菌有高度傳遞抗藥基因至其他細菌的能力，造成毒性強大的細菌在藥物治療上的困難，也是引發院內感染的重要細菌。然而，現在在臨床在檢驗藥敏實驗上，不僅昂貴、消耗人力資源還耗時，會延誤隔離時間增加院內傳染的機會。相比之下，現在臨床中檢驗菌株的方法:基質輔助激光解吸/電離是一個快速且高效率的工具。
在此研究中，我們透過機器學習的方法分析基質輔助激光解吸/電離工具中所得到的質譜圖來預測腸球菌對萬古黴素的抗藥性，讓我們能更提前預防VRE的威脅。在此研究中，獲得2922張對萬古黴素無抗藥性和2795張對萬古黴素有抗藥性之質譜圖，透過統計方法找出所有圖譜中能分辨有無抗藥性的PEAK。再透過機器學習分類器演算法和交叉驗證訓練的實驗，測試出適合的資料處理方法、演算法和特徵屬性來建構預測模型。同時使用獨立資料來測試模型的可信度，最後模型能分辨腸球菌有無抗藥性的準確率為八成左右，也特別找到再m/z為6690,6603,3302,3165,7360等峰值是有分辨VRE的關鍵峰值。經由大量的資料和嚴格的驗證方法，本研究中的預測模型能提供臨床醫師一個可信的方向，來提前隔離預防細菌的傳染以及抗藥性的轉移。

Vancomycin-resistant Enterococcus(VRE) is a common Gram-positive coccus with implicit antimicrobial resistance to vancomycin, and may cause the urinary tract infections, bacteremia, endocarditis infections. In 1985, the possibility of transfer of vancomycin resistance genes to other gram-positive was discovered, causing difficulties in drug treatment of highly toxic bacteria and increasing the risk of nosocomial infections. However, the antimicrobial susceptibility experiments in clinical trials is an expensive and time-consuming method, which delay the isolation time. In contrast, matrix-assisted laser desorption ionization time-of-flight (MALDI-TOF), bacterial identification method, is a high efficiency tool to distinguish antibiotics resistance.
In this study, we developed the machine learning model to analyzed MALDI-TOF mass spectra about Enterococcus and predict vancomycin-resistance. Mass spectra were collected from 2922 vancomycin-susceptible Enterococcus(VSE) and 2795 vancomycin-resistant Enterococcus(VRE). A total of 100 discriminative features were selected from all mass spectrums by evaluating model performance. The independent data set was used to test the credibility of the model. The area under the curve of random forest model on discriminating vancomycin-resistant reached 0.855 and the accuracy was about 80%. The predictive model provides clinicians with a credible direction to isolate bacterial infections and antimicrobial resistance in advance to decrease the risk of vancomycin resistance genes transfer.

摘要 iii
ABSTRACT iv
致謝 vi
Contents vii
List of Tables ix
List of figures x
Chapter 1 – Introduction 1
1.1 Background 1
1.2 Motivation and Goal 2
1.3 Relative Work 3
Chapter 2 - Material 6
2.1 Collection of MALDI-TOF MS spectra for E. faecium 6
2.2 Vancomycin susceptibility testing of E. faecium 6
2.3 Construction of Training and Testing data sets 7
Chapter 3 – Methods 10
3.1 Study Design 10
3.2 Data preprocessing 11
3.2.1 Align raw data to internal calibrates 11
3.3 Feature Extraction and Selection 13
3.3.1 Binning method 13
3.3.2 Kernel Density Estimation-Gaussian distribution 15
3.3.3 Chi-square test for selecting identification peak 18
3.4Models training 19
3.4.1 Random forest 19
3.4.2 Support vector machine 20
3.4.3 K-nearest neighbor 21
3.5 Performance evaluation 22
3.5.1 five-fold cross-validation 23
3.5.2 time-wise internal validation 23
3.5.3 independent testing for external validation 24
3.5.4 Performance metrics 24
Chapter 4 – Result 26
4.1 Align to specific biomarker peak m/z 4492 26
4.2 The total of 100 predictive peaks for detecting VREfm 28
4.2.1 Appropriate bin size solve the peak shifting problem 28
4.2.2 Precise m/z value calculated by KDE method 30
4.3 Selected m/z peaks for detecting VREfm 33
4.4 Performance of the Predictive Models 36
4.5 Web server tool 40
Chapter 5 – Summary 43
5.1 Discussion 43
5.2 Identification of discriminating peaks 44
5.3 Conclusions 49
Reference 50

1. Fischetti, V.A. and American Society for Microbiology., Gram-positive pathogens. 2nd ed. 2006, Washington, D.C.: ASM Press. xiii, 849 p., 22 p. of plates.
2. Fisher, K. and C. Phillips, The ecology, epidemiology and virulence of Enterococcus. Microbiology, 2009. 155(Pt 6): p. 1749-57.
3. Zhanel, G.G., D.J. Hoban, and J.A. Karlowsky, Nitrofurantoin is active against vancomycin-resistant enterococci. Antimicrob Agents Chemother, 2001. 45(1): p. 324-6.
4. Uttley, A.H., et al., Vancomycin-resistant enterococci. Lancet, 1988. 1(8575-6): p. 57-8.
5. Ott, E., et al., The prevalence of nosocomial and community acquired infections in a university hospital: an observational study. Dtsch Arztebl Int, 2013. 110(31-32): p. 533-40.
6. Gastmeier, P., et al., Dramatic increase in vancomycin-resistant enterococci in Germany. J Antimicrob Chemother, 2014. 69(6): p. 1660-4.
7. Hegstad, K., et al., Mobile genetic elements and their contribution to the emergence of antimicrobial resistant Enterococcus faecalis and Enterococcus faecium. Clin Microbiol Infect, 2010. 16(6): p. 541-54.
8. Cetinkaya, Y., P. Falk, and C.G. Mayhall, Vancomycin-resistant enterococci. Clin Microbiol Rev, 2000. 13(4): p. 686-707.
9. de Niederhausern, S., et al., Vancomycin-resistance transferability from VanA enterococci to Staphylococcus aureus. Curr Microbiol, 2011. 62(5): p. 1363-7.
10. Perichon, B. and P. Courvalin, VanA-type vancomycin-resistant Staphylococcus aureus. Antimicrob Agents Chemother, 2009. 53(11): p. 4580-7.
11. Bourgeois-Nicolaos, N., et al., Comparative study of vanA gene transfer from Enterococcus faecium to Enterococcus faecalis and to Enterococcus faecium in the intestine of mice. FEMS Microbiol Lett, 2006. 254(1): p. 27-33.
12. Sievert, D.M., et al., Antimicrobial-resistant pathogens associated with healthcare-associated infections: summary of data reported to the National Healthcare Safety Network at the Centers for Disease Control and Prevention, 2009-2010. Infect Control Hosp Epidemiol, 2013. 34(1): p. 1-14.
13. Lester, C.H., et al., Emergence of ampicillin-resistant Enterococcus faecium in Danish hospitals. J Antimicrob Chemother, 2008. 62(6): p. 1203-6.
14. Sader, H.S., et al., Antimicrobial susceptibility of daptomycin and comparator agents tested against methicillin-resistant Staphylococcus aureus and vancomycin-resistant enterococci: trend analysis of a 6-year period in US medical centers (2005-2010). Diagn Microbiol Infect Dis, 2011. 70(3): p. 412-6.
15. Lodise, T.P., et al., Clinical outcomes for patients with bacteremia caused by vancomycin-resistant enterococcus in a level 1 trauma center. Clin Infect Dis, 2002. 34(7): p. 922-9.
16. Ghanem, G., et al., Outcomes for and risk factors associated with vancomycin-resistant Enterococcus faecalis and vancomycin-resistant Enterococcus faecium bacteremia in cancer patients. Infect Control Hosp Epidemiol, 2007. 28(9): p. 1054-9.
17. Ge, M.C., et al., Routine identification of microorganisms by matrix-assisted laser desorption ionization time-of-flight mass spectrometry: Success rate, economic analysis, and clinical outcome. J Microbiol Immunol Infect, 2017. 50(5): p. 662-668.
18. Vrioni, G., et al., MALDI-TOF mass spectrometry technology for detecting biomarkers of antimicrobial resistance: current achievements and future perspectives. Ann Transl Med, 2018. 6(12): p. 240.
19. Singhal, N., et al., MALDI-TOF mass spectrometry: an emerging technology for microbial identification and diagnosis. Front Microbiol, 2015. 6: p. 791.
20. Carbonnelle, E., et al., MALDI-TOF mass spectrometry tools for bacterial identification in clinical microbiology laboratory. Clin Biochem, 2011. 44(1): p. 104-9.
21. Wang, H.Y., et al., Application of a MALDI-TOF analysis platform (ClinProTools) for rapid and preliminary report of MRSA sequence types in Taiwan. PeerJ, 2018. 6: p. e5784.
22. Lu, J.J., et al., Peptide biomarker discovery for identification of methicillin-resistant and vancomycin-intermediate Staphylococcus aureus strains by MALDI-TOF. Anal Chem, 2012. 84(13): p. 5685-92.
23. Mellmann, A., et al., High interlaboratory reproducibility of matrix-assisted laser desorption ionization-time of flight mass spectrometry-based species identification of nonfermenting bacteria. J Clin Microbiol, 2009. 47(11): p. 3732-4.
24. Wang, H.Y., et al., Rapid Detection of Heterogeneous Vancomycin-Intermediate Staphylococcus aureus Based on Matrix-Assisted Laser Desorption Ionization Time-of-Flight: Using a Machine Learning Approach and Unbiased Validation. Front Microbiol, 2018. 9: p. 2393.
25. Nomura, F., Proteome-based bacterial identification using matrix-assisted laser desorption ionization-time of flight mass spectrometry (MALDI-TOF MS): A revolutionary shift in clinical diagnostic microbiology. Biochim Biophys Acta, 2015. 1854(6): p. 528-37.
26. Lo-Ciganic, W.H., et al., Evaluation of Machine-Learning Algorithms for Predicting Opioid Overdose Risk Among Medicare Beneficiaries With Opioid Prescriptions. JAMA Netw Open, 2019. 2(3): p. e190968.
27. Tseng, Y.J., et al., Predicting breast cancer metastasis by using serum biomarkers and clinicopathological data with machine learning technologies. Int J Med Inform, 2019.
28. Kuppermann, N., et al., A Clinical Prediction Rule to Identify Febrile Infants 60 Days and Younger at Low Risk for Serious Bacterial Infections. JAMA Pediatr, 2019.
29. Karter, A.J., et al., Development and Validation of a Tool to Identify Patients With Type 2 Diabetes at High Risk of Hypoglycemia-Related Emergency Department or Hospital Use. JAMA Intern Med, 2017. 177(10): p. 1461-1470.
30. Gulshan, V., et al., Development and Validation of a Deep Learning Algorithm for Detection of Diabetic Retinopathy in Retinal Fundus Photographs. JAMA, 2016. 316(22): p. 2402-2410.
31. Dever, L.A. and T.S. Dermody, Mechanisms of bacterial resistance to antibiotics. Arch Intern Med, 1991. 151(5): p. 886-95.
32. Griffin, P.M., et al., Use of matrix-assisted laser desorption ionization-time of flight mass spectrometry to identify vancomycin-resistant enterococci and investigate the epidemiology of an outbreak. J Clin Microbiol, 2012. 50(9): p. 2918-31.
33. Lasch, P., et al., Insufficient discriminatory power of MALDI-TOF mass spectrometry for typing of Enterococcus faecium and Staphylococcus aureus isolates. J Microbiol Methods, 2014. 100: p. 58-69.
34. Norris, J.L., et al., Processing MALDI Mass Spectra to Improve Mass Spectral Direct Tissue Analysis. Int J Mass Spectrom, 2007. 260(2-3): p. 212-221.
35. Cannataro, M., et al., Preprocessing of Mass Spectrometry Proteomics Data on the Grid, in 18th IEEE Symposium on Computer-Based Medical Systems (CBMS'05). 2005. p. 549-554.
36. Kazmi, S.A., et al., Alignment of high resolution mass spectra: development of a heuristic approach for metabolomics. Metabolomics, 2006. 2(2): p. 75-83.
37. Alexandrov, T., MALDI imaging mass spectrometry: statistical data analysis and current computational challenges. BMC Bioinformatics, 2012. 13(S16).
38. Becker, K., et al., A new scheme for strain typing of methicillin-resistant Staphylococcus aureus on the basis of matrix-assisted laser desorption ionization time-of-flight mass spectrometry by using machine learning approach. Plos One, 2018. 13(3).
39. Rosenblatt, M., Remarks on Some Nonparametric Estimates of a Density Function. The Annals of Mathematical Statistics, 1956. 27(3): p. 832-837.
40. Parzen, E., On Estimation of a Probability Density Function and Mode. The Annals of Mathematical Statistics, 1962. 33(3): p. 1065-1076.
41. Pearson, K., X. On the criterion that a given system of deviations from the probable in the case of a correlated system of variables is such that it can be reasonably supposed to have arisen from random sampling. The London, Edinburgh, and Dublin Philosophical Magazine and Journal of Science, 2009. 50(302): p. 157-175.
42. Hall, M., et al., The WEKA data mining software. ACM SIGKDD Explorations Newsletter, 2009. 11(1).
43. Breiman, L., Machine Learning, 2001. 45(1): p. 5-32.
44. Cortes, C. and V. Vapnik, Support-vector networks. Machine Learning, 1995. 20(3): p. 273-297.
45. Vapnik, V.N., An overview of statistical learning theory. IEEE Transactions on Neural Networks, 1999. 10(5): p. 988-999.
46. Byvatov, E. and G. Schneider, Support vector machine applications in bioinformatics. Appl Bioinformatics, 2003. 2(2): p. 67-77.
47. Kumari, B., R. Kumar, and M. Kumar, PalmPred: an SVM based palmitoylation prediction method using sequence profile information. PLoS One, 2014. 9(2): p. e89246.
48. Lu, C.T., et al., Carboxylator: incorporating solvent-accessible surface area for identifying protein carboxylation sites. J Comput Aided Mol Des, 2011. 25(10): p. 987-95.
49. Lee, T.Y., et al., Incorporating distant sequence features and radial basis function networks to identify ubiquitin conjugation sites. PLoS One, 2011. 6(3): p. e17331.
50. Chang, W.C., et al., Incorporating support vector machine for identifying protein tyrosine sulfation sites. J Comput Chem, 2009. 30(15): p. 2526-37.
51. Wong, Y.H., et al., KinasePhos 2.0: a web server for identifying protein kinase-specific phosphorylation sites based on sequences and coupling patterns. Nucleic Acids Res, 2007. 35(Web Server issue): p. W588-94.
52. Chang, C.-C. and C.-J. Lin, Libsvm. ACM Transactions on Intelligent Systems and Technology, 2011. 2(3): p. 1-27.
53. Varma, S. and R. Simon, Bias in error estimation when using cross-validation for model selection. BMC Bioinformatics, 2006. 7: p. 91.
54. Krstajic, D., et al., Cross-validation pitfalls when selecting and assessing regression and classification models. J Cheminform, 2014. 6(1): p. 10.
55. Quintela-Baluja, M., et al., Characterization of different food-isolatedEnterococcusstrains by MALDI-TOF mass fingerprinting. Electrophoresis, 2013. 34(15): p. 2240-2250.
56. Holzknecht, B.J., et al., Typing of vancomycin-resistant enterococci with MALDI-TOF mass spectrometry in a nosocomial outbreak setting. Clin Microbiol Infect, 2018. 24(10): p. 1104 e1-1104 e4.
57. Hrabak, J., E. Chudackova, and R. Walkova, Matrix-assisted laser desorption ionization-time of flight (maldi-tof) mass spectrometry for detection of antibiotic resistance mechanisms: from research to routine diagnosis. Clin Microbiol Rev, 2013. 26(1): p. 103-14.
58. Idelevich, E.A., et al., Rapid detection of antibiotic resistance by MALDI-TOF mass spectrometry using a novel direct-on-target microdroplet growth assay. Clin Microbiol Infect, 2018. 24(7): p. 738-743.
59. Suarez, S., et al., Ribosomal proteins as biomarkers for bacterial identification by mass spectrometry in the clinical microbiology laboratory. J Microbiol Methods, 2013. 94(3): p. 390-6.
60. Wang, H.Y., et al., A new scheme for strain typing of methicillin-resistant Staphylococcus aureus on the basis of matrix-assisted laser desorption ionization time-of-flight mass spectrometry by using machine learning approach. PLoS One, 2018. 13(3): p. e0194289.
61. Burckhardt, I. and S. Zimmermann, Susceptibility Testing of Bacteria Using Maldi-Tof Mass Spectrometry. Front Microbiol, 2018. 9: p. 1744.
62. Kostrzewa, M., et al., MALDI-TOF MS: an upcoming tool for rapid detection of antibiotic resistance in microorganisms. Proteomics Clin Appl, 2013. 7(11-12): p. 767-78.
63. Mather, C.A., et al., Rapid Detection of Vancomycin-Intermediate Staphylococcus aureus by Matrix-Assisted Laser Desorption Ionization-Time of Flight Mass Spectrometry. J Clin Microbiol, 2016. 54(4): p. 883-90.
64. Rhie, K., et al., Etiology of Invasive Bacterial Infections in Immunocompetent Children in Korea (2006–2010): a Retrospective Multicenter Study. Journal of Korean Medical Science, 2018. 33(6).
65. Miller, W.R., J.M. Munita, and C.A. Arias, Mechanisms of antibiotic resistance in enterococci. Expert Review of Anti-infective Therapy, 2014. 12(10): p. 1221-1236.
66. Lebreton, F., et al., D-Ala-d-Ser VanN-type transferable vancomycin resistance in Enterococcus faecium. Antimicrob Agents Chemother, 2011. 55(10): p. 4606-12