Biomarker Selection and a Prospective Metabolite-based Machine Learning Diagnostic for Lyme Disease

Overview

Journal Sci Rep

Specialty Science

Date 2022 Jan 28

PMID 35087163

Authors

Eric R Kehoe

Bryna L Fitzgerald

Barbara Graham

M Nurul Islam

Kartikay Sharma

Gary P Wormser

John T Belisle

Michael J Kirby

Affiliations

Soon will be listed here.

Abstract

We provide a pipeline for data preprocessing, biomarker selection, and classification of liquid chromatography-mass spectrometry (LCMS) serum samples to generate a prospective diagnostic test for Lyme disease. We utilize tools of machine learning (ML), e.g., sparse support vector machines (SSVM), iterative feature removal (IFR), and k-fold feature ranking to select several biomarkers and build a discriminant model for Lyme disease. We report a 98.13% test balanced success rate (BSR) of our model based on a sequestered test set of LCMS serum samples. The methodology employed is general and can be readily adapted to other LCMS, or metabolomics, data sets.

Citing Articles

Artificial Intelligence in Metabolomics: A Current Review.

Chi J, Shu J, Li M, Mudappathi R, Jin Y, Lewis F Trends Analyt Chem. 2024; 178.

PMID: 39071116 PMC: 11271759. DOI: 10.1016/j.trac.2024.117852.

Lyme rashes disease classification using deep feature fusion technique.

Ali G, Anwar M, Nauman M, Faheem M, Rashid J Skin Res Technol. 2023; 29(11):e13519.

PMID: 38009027 PMC: 10628356. DOI: 10.1111/srt.13519.

Wearable chemical sensors for biomarker discovery in the omics era.

Sempionatto J, Lasalde-Ramirez J, Mahato K, Wang J, Gao W Nat Rev Chem. 2023; 6(12):899-915.

PMID: 37117704 PMC: 9666953. DOI: 10.1038/s41570-022-00439-w.

Using machine learning to determine the time of exposure to infection by a respiratory pathogen.

Sharma K, Aminian M, Ghosh T, Liu X, Kirby M Sci Rep. 2023; 13(1):5340.

PMID: 37005391 PMC: 10067823. DOI: 10.1038/s41598-023-30306-7.

Precision Medicine Approaches with Metabolomics and Artificial Intelligence.

Barberis E, Khoso S, Sica A, Falasca M, Gennari A, Dondero F Int J Mol Sci. 2022; 23(19).

PMID: 36232571 PMC: 9569627. DOI: 10.3390/ijms231911269.

References

Dieterle F, Ross A, Schlotterbeck G, Senn H . Probabilistic quotient normalization as robust method to account for dilution of complex biological mixtures. Application in 1H NMR metabonomics. Anal Chem. 2006; 78(13):4281-90. DOI: 10.1021/ac051632c. View

Chambers M, MacLean B, Burke R, Amodei D, Ruderman D, Neumann S . A cross-platform toolkit for mass spectrometry and proteomics. Nat Biotechnol. 2012; 30(10):918-20. PMC: 3471674. DOI: 10.1038/nbt.2377. View

Mahadevan S, Shah S, Marrie T, Slupsky C . Analysis of metabolomic data using support vector machines. Anal Chem. 2008; 80(19):7562-70. DOI: 10.1021/ac800954c. View

Ghosh T, Zhang W, Ghosh D, Kechris K . Predictive Modeling for Metabolomics Data. Methods Mol Biol. 2020; 2104:313-336. PMC: 7423323. DOI: 10.1007/978-1-0716-0239-3_16. View

Veselkov K, Vingara L, Masson P, Robinette S, Want E, Li J . Optimized preprocessing of ultra-performance liquid chromatography/mass spectrometry urinary metabolic profiles for improved information recovery. Anal Chem. 2011; 83(15):5864-72. DOI: 10.1021/ac201065j. View

Donoho D, Tanner J . Sparse nonnegative solution of underdetermined linear equations by linear programming. Proc Natl Acad Sci U S A. 2005; 102(27):9446-51. PMC: 1172251. DOI: 10.1073/pnas.0502269102. View

OHara S, Wang K, Slayden R, Schenkel A, Huber G, OHern C . Iterative feature removal yields highly discriminative pathways. BMC Genomics. 2013; 14:832. PMC: 3879090. DOI: 10.1186/1471-2164-14-832. View

Guan W, Zhou M, Hampton C, Benigno B, Walker L, Gray A . Ovarian cancer detection from metabolomic liquid chromatography/mass spectrometry data by support vector machines. BMC Bioinformatics. 2009; 10:259. PMC: 2741455. DOI: 10.1186/1471-2105-10-259. View

Molins C, Ashton L, Wormser G, Hess A, Delorey M, Mahapatra S . Development of a metabolic biosignature for detection of early Lyme disease. Clin Infect Dis. 2015; 60(12):1767-75. PMC: 4810808. DOI: 10.1093/cid/civ185. View

10.

Lee J, Styczynski M . NS-kNN: a modified k-nearest neighbors approach for imputing metabolomics data. Metabolomics. 2019; 14(12):153. PMC: 6532628. DOI: 10.1007/s11306-018-1451-8. View

11.

Lee L, Liong C, Jemain A . Partial least squares-discriminant analysis (PLS-DA) for classification of high-dimensional (HD) data: a review of contemporary practice strategies and knowledge gaps. Analyst. 2018; 143(15):3526-3539. DOI: 10.1039/c8an00599k. View

12.

Schymanski E, Jeon J, Gulde R, Fenner K, Ruff M, Singer H . Identifying small molecules via high resolution mass spectrometry: communicating confidence. Environ Sci Technol. 2014; 48(4):2097-8. DOI: 10.1021/es5002105. View

13.

Steere A, Strle F, Wormser G, Hu L, Branda J, Hovius J . Lyme borreliosis. Nat Rev Dis Primers. 2016; 2:16090. PMC: 5539539. DOI: 10.1038/nrdp.2016.90. View

14.

Prince J, Marcotte E . Chromatographic alignment of ESI-LC-MS proteomics data sets by ordered bijective interpolated warping. Anal Chem. 2006; 78(17):6140-52. DOI: 10.1021/ac0605344. View

15.

Hawkins D . The problem of overfitting. J Chem Inf Comput Sci. 2004; 44(1):1-12. DOI: 10.1021/ci0342472. View

16.

Pegalajar-Jurado A, Fitzgerald B, Islam M, Belisle J, Wormser G, Waller K . Identification of Urine Metabolites as Biomarkers of Early Lyme Disease. Sci Rep. 2018; 8(1):12204. PMC: 6093930. DOI: 10.1038/s41598-018-29713-y. View

17.

Kerstholt M, Vrijmoeth H, Lachmandas E, Oosting M, Lupse M, Flonta M . Role of glutathione metabolism in host defense against infection. Proc Natl Acad Sci U S A. 2018; 115(10):E2320-E2328. PMC: 5877983. DOI: 10.1073/pnas.1720833115. View

18.

Tautenhahn R, Bottcher C, Neumann S . Highly sensitive feature detection for high resolution LC/MS. BMC Bioinformatics. 2008; 9:504. PMC: 2639432. DOI: 10.1186/1471-2105-9-504. View

19.

Alakwaa F, Chaudhary K, Garmire L . Deep Learning Accurately Predicts Estrogen Receptor Status in Breast Cancer Metabolomics Data. J Proteome Res. 2017; 17(1):337-347. PMC: 5759031. DOI: 10.1021/acs.jproteome.7b00595. View

20.

Smith C, Want E, OMaille G, Abagyan R, Siuzdak G . XCMS: processing mass spectrometry data for metabolite profiling using nonlinear peak alignment, matching, and identification. Anal Chem. 2006; 78(3):779-87. DOI: 10.1021/ac051437y. View