Comparing biomarkers and proteomic fingerprints for classification studies

Abstract

Early disease detection is extremely important in the treatment and prognosis of many diseases, especially cancer. Often, proteomic fingerprints and a pattern recognition algorithm are used to classify the pathological condition of a given individual. It has been argued that accurate classification of the existing data implies an underlying biological significance. Two fingerprint-based classifiers, decision tree and medoid classification algorithm, and a biomarker-based classifier were examined using a published dataset of mass spectral peaks from 81 healthy individuals and 78 individuals with benign prostate hyperplasia (BPH). For all three methods, classifiers were constructed using the original data and the data after permuting the labels of the samples (BPH and healthy). The fingerprint-based classifiers produced accurate results for the original data, though the peaks used in a given classifier depended upon which samples were placed in the training set. Accurate results were also obtained for the dataset with permuted labels. In contrast, only three unique peaks were identified as putative biomarkers, producing a small number of reasonably accurate biomarker-based classifiers. The dataset with permuted labels was poorly classified. Since fingerprint-based classifiers accurately classified the dataset with permuted labels, the argument for biological significance from a fingerprint-based classifier must be questioned.

Share and Cite:

Luke, B. , Collins, J. , Habermann, J. , Prieto, D. , Veenstra, T. and Ried, T. (2013) Comparing biomarkers and proteomic fingerprints for classification studies. Journal of Biomedical Science and Engineering, 6, 453-465. doi: 10.4236/jbise.2013.64057.

Conflicts of Interest

The authors declare no conflicts of interest.

References

[1] Ho, D.W., Yang, Z.F., Wong, B.Y., Kwong, D.L., Sham, J.S., Wei, W.I., et al. (2006) Surface-enhanced laser desorption/ionization time-of-flight mass spectrometry serum protein profiling to identify nasopharyngeal carcinoma. Cancer, 107, 99-107. doi:10.1002/cncr.21970
[2] Yu, Y., Chen, S., Wang, L.S., Chen, W.L., Guo, W.J., Yan, H., et al. (2005) Prediction of pancreatic cancer by serum biomarkers using surface-enhanced laser desorption/ionization-based decision tree classification. Oncology, 68, 79-86. doi:10.1159/000084824
[3] Yang, S.Y., Xiao, X.Y., Zhang, W.G., Zhang, L.J., Zhang, W., Zhou, B., et al. (2005) Application of serum SELDI proteomic patterns in diagnosis of lung cancer. BMC Cancer, 5, 83. doi:10.1186/1471-2407-5-83
[4] Liu, W., Guan, M., Wu, D., Zhang, Y., Wu, Z., Xu, M. and Lu, Y. (2005) Using tree analysis pattern and SE- LDI-TOF-MS to discriminate transitional cell carcinoma of the bladder cancer from noncancer patients. European Urology, 47, 456-462. doi:10.1016/j.eururo.2004.10.006
[5] Srinivasan, R., Daniels, J., Fusaro, V., Lundqvist, A., Killian, J.K., Geho, D., et al. (2006) Accurate diagnosis of acute graft-versus-host disease using serum proteomic pattern analysis. Experimental Hematology, 34, 796-801. doi:10.1016/j.exphem.2006.02.013
[6] Stone, J.H., Rajapakse, V.N., Hoffman, G.S., Specks, U., Merkel, P.A., Spiera, R.F., et al. (2005) A serum proteomic approach to gauging the state of remission in Wegener’s granulomatosis. Arthritis & Rheumatism, 52, 902- 910. doi:10.1002/art.20938
[7] Brouwers, F.M., Petricoin 3rd, E.F., Ksinantova, L., Breza, J., Rajapakse, V., Ross, S., et al. (2005) Low molecular weight proteomic information distinguishes metastatic from benign pheochromocytoma. Endocrine-Related Cancer, 12, 263-272. doi:10.1677/erc.1.00913
[8] Petricoin, E.F., Rajapaske, V., Herman, E.H., Arekani, A.M., Ross, S., Johann, D., et al. (2004) Toxico-proteomics: Serum proteomic pattern diagnostics for early detection of drug induced cardiac toxicities and cardioprotection. Toxicologic Pathology, 32, 122-130. doi:10.1080/01926230490426516
[9] Ornstein, D.K., Rayford, W., Fusaro, V.A., Conrads, T.P., Ross, S.J., Hitt, B.A., et al. (2004) Serum proteomic profiling can discriminate prostate cancer from benign prostates in men with total prostate specific antigen levels between 2.5 and 15.0 ng/ml. Journal of Urology, 172, 1302- 1305. doi:10.1097/01.ju.0000139572.88463.39
[10] Conrads, T.P., Fusaro, V.A., Ross, S., Johann, D., Rajapakse, V., Hitt, B.A., et al. (2004) High-resolution serum proteomic features for ovarian cancer detection. Endocrine-Related Cancer, 11, 163-178. doi:10.1677/erc.0.0110163
[11] Petricoin, E.F., Fishman, D.A., Conrads, T.P., Veenstra, T.D. and Liotta, L.A. (2003) Proteomic pattern diagnostics: Producers and consumers in the era of correlative science. BMC Bioinformatics, 4, 24. doi:10.1186/1471-2105-4-24
[12] Adam, B.L., Qu, Y., Davis, J.W., Ward, M.D., Clements, M.A., Cazares, L.H., et al. (2002) Serum protein fingerprinting coupled with a pattern-matching algorithm distinguishes prostate cancer from benign prostate hyperplasia and healthy men. Cancer Research, 62, 3609-3614.
[13] The BioMarker Discovery Kit. http://isp.ncifcrf.gov/abcc/abcc-groups/simulation-and-modeling/biomarker-discovery-kit/
[14] Habermann, J.K., Roblick, U.J., Luke, B.T., Prieto, D.A., Finlay, W.J., Podust, V.N., et al. (2006) Increased serum levels of complement C3a anaphylatoxin indicate the presence of colorectal tumors. Gastroenterology, 131, 1020- 1029. doi:10.1053/j.gastro.2006.07.011
[15] Luke, B.T. (2003) Genetic algorithms and beyond. In: Leardi, R., Ed., Nature-Inspired Methods in Chemometrics: Genetic Algorithms and Neural Networks, Chapter 12. Elsevier, Amsterdam. doi:10.1016/S0922-3487(03)23001-X
[16] Ransohoff, D.F. (2005) Lessons from controversy: Ovarian cancer screening and serum proteomics. Journal of the National Cancer Institute, 97, 315-319. doi:10.1093/jnci/dji054
[17] Ransohoff, D.F. (2005) Bias as a threat to the validity of cancer molecular-marker research. Nature Reviews Cancer, 5, 142-149. doi:10.1038/nrc1550
[18] Wulfkuhle, J.D., Edmiston, K.H., Liotta, L.A. and Petricoin 3rd, E.F. (2006) Technology insight: Pharmaco-proteomics for cancer—Promises of patient-tailored medicine using protein microarrays. Nature Clinical Practice Oncology, 3, 256-268. doi:10.1038/ncponc0485
[19] Gulmann, C., Sheehan, K.M., Kay, E.W., Liotta, L.A., and Petricoin 3rd, E.F. (2006) Array-based proteomics: Mapping of protein circuitries for diagnostics, prognostics, and therapy guidance in cancer. Journal of Pathology, 208, 595-606. doi:10.1002/path.1958
[20] Petricoin 3rd, E.F., Bichsel, V.E., Calvert, V.S., Espina, V., Winters, M., Young, L., et al. (2005) Mapping molecular networks using proteomics: A vision for patienttailored combination therapy. Journal of Clinical Oncology, 23, 3614-3621. doi:10.1200/JCO.2005.02.509
[21] Espina, V., Geho, D., Mehta, A.I., Petricoin 3rd, E.F., Liotta, L.A. and Rosenblatt, K.P. (2005) Pathology of the future: Molecular profiling for targeted therapy. Cancer Investigation, 23, 36-46. doi:10.1081/CNV-46434
[22] Calvo, K.R., Liotta, L.A. and Petricoin, E.F. (2005) Clinical proteomics: From biomarker discovery and cell signaling profiles to individualized personal therapy. Bioscience Reports, 25, 107-125. doi:10.1007/s10540-005-2851-3
[23] Anderson, N.L. and Anderson, N.G. (2002) The human plasma proteome: History, character, and diagnostic prospects. Molecular & Cellular Proteomics, 1, 845-867. doi:10.1074/mcp.R200007-MCP200

Copyright © 2024 by authors and Scientific Research Publishing Inc.

Creative Commons License

This work and the related PDF file are licensed under a Creative Commons Attribution 4.0 International License.