Multivariate Analysis for Metabolomics and Proteomics Data - Proteomic and Metabolomic Approaches to Biomarker Discovery - page 303

key spectral regions identi ed in the PLS-DA

analysis in patients without CAD, which substan-
tially reduces the predictive power of the model.
Thus the

6. Fisher RA. The use of multiple measurements in taxo-
nomic problems. Ann Eugen 1936; 7 :179 e 88.
7. Wikipedia Iris ower data set. (n.d.). Retrieved May 2012

from Wikipedia: http://en.wikipedia.org/wiki/Iris_

final lesson from this case study is to
beware of confounding effects, as they can be
very dif

ower_data_set .
8. Esbensen K. Multivariate Data Analysis in Practice .
CAMO Process AS; 2002.
9. Wold S, Josefson M. Multivariate analysis of analytical
data. In: Encyclopedia of Analytical Chemistry, edited by
Meyers, RA . John Wiley & Sons; 2000. p. 9710 e 36.
10. Balch CM, Buzaid AC, Soong SJ, et al. Final version of
the American Joint Committee on Cancer staging system
for cutaneous melanoma. JClinOncol 2001; 19 :3635 e 48.
11. Caron J, Mangé A, Guillot B, et al. Highly sensitive
detection of melanoma based on serum proteomic
pro

cult to identify with purely statistical
methods, and domain knowledge is key to avoid-
ing them.

CONCLUSIONS

The large amount of data that modern proteo-
mic and metabolomic experiments provide is
a challenge to researchers. MVA is an essential
tool for the analysis and interpretation of data
frommodern metabolomic and proteomic exper-
iments. This chapter is a brief summary of the two
essential methods of MVA: principal component
analysis and partial least squares. Two recent
examples from the life science literature are dis-
cussed. There is not suf

ling. J Cancer Res Clin Oncol 2009; 135 :1257 e 64.
12. Roberts R, Stewart AF. The genetics of coronary artery
disease. Curr Opin Cardiol 2012; 27 :221 e 7.
13. Brindle JT, Antti H, Holmes E, et al. Rapid and
noninvasive diagnosis of the presence and severity of
coronary heart disease using 1 H-NMR-based metabo-
nomics. Nat Med 2002; 12 :1439 e 44.
14. Kirschenlohr HL, Grif n JL, Clarke SC, et al. Proton

NMR analysis of plasma is a weak predictor of coro-
nary artery disease. Nat Med 2006; 12 :705 e 10.
15. Enot DP, Haas B, Weinberger KM. Bioinformatics for
mass spectrometry-based metabolomics. Methods Mol
Biol 2011; 719 :351 e 75.
16. Noble WS, MacCoss MJ. Computational and statistical
analysis of protein mass spectrometry data. PLOS
Comput Biol 2012; 8 :e1002296.
17. Gadbury GL, Garrett KA, Allison DB. Challenges and
approaches to statistical design and inference in high-
dimensional investigations. Methods Mol Biol 2009; 553 :
181 e 206.
18. Kholodenko B, Yaffe MB, Kolch W. Computational
approaches for analyzing information

cient space in this
chapter to discuss or even overview all of the
numerous techniques and methods that have
been used in this area in recent years. The inter-
ested reader is recommended to consult the
following references for more methods and case
studies of data analysis of metabolomic and pro-
teomic experiments. 15 e 22

flow in biological

References
1. Allegra CJ, Jessup JM, Somer

networks. Sci Signal 2012; 5 (220): re1.
19. Gebregiworgis T, Powers R. Application of NMR
metabolomics to search for human disease biomarkers.
Comb Chem High Throughput Screen 2012; 15 :595 e 610.
20. Gu H, Pan Z, Xi B, Asiago V, Musselman B, Raftery D.
Principal component directed partial least squares
analysis for combining nuclear magnetic resonance and
mass spectrometry data in metabolomics: application to
the detection of breast cancer. Anal Chim Acta 2011; 686 :
57 e 63.
21. Madsen R, Lundstedt T, Trygg J. Chemometrics in
metabolomics d a review in human disease diagnosis.
Anal Chim Acta 2010; 659 :23 e 33.
22. Reshef DN, Reshef YA, Finucane HK, et al. Detecting
novel associations in large data sets. Science 2011; 334 :
1518 e 24.

eld MR, et al. American
Society of Clinical Oncology provisional clinical opinion:
testing for KRAS gene mutations in patients with meta-
static colorectal carcinoma to predict response to anti-
epidermal growth factor receptor monoclonal antibody
therapy. J Clin Oncol 2009; 27 :2091 e 6.
2. Hamburg MA, Collins FA. The path to personalized
medicine. N Engl J Med 2010; 363 :301 e 4.
3. van t Veer LJ, Dai H, van de Vijver MJ, et al. Gene

expression pro

'

ling predicts clinical outcome of breast
cancer. Nature 2002; 415 :530 e 6.
4. Eriksson L, Johansson E, Kettaneh-WoldN, Trygg J, Wold S.
Multi-Megavariate Data Analysis. Umetrics AB ; 2006.

5. Strang G. Computational Science and Engineering .

Wellesley-Cambridge Press; 2007.