Biology Reference
In-Depth Information
CHAPTER
19
Multivariate Analysis for Metabolomics
a nd Proteomics Da ta
Frederick H. Long
Spectroscopic Solutions, LLC, Randolph, NJ, USA, Henry W. Long Center for Functional Cancer
Epigenetics and Dana-Faber Cancer Institute and Harvard Medical School, Boston, MA, USA
OUTLINE
Study 1: Cancer Detection by Proteomics
305
Conclusions
311
Study 2: Detection of Heart Disease by
Metabolomics
References
311
306
global evaluation 3 using Affymetrix technology.
In most cases, proteins are the end product of
transcribed DNA and represent the key molec-
ular machines and building blocks of cells.
Proteins and metabolites are measured by the
spectroscopic analysis of
To understand the many approaches to iden-
tifying biomarkers, it helps to consider the so-
called central dogma of molecular biology
(DNA codes for RNA, which codes for proteins
that act on other proteins or small molecules).
Current technology as of this writing (mid-
2012) allows for the complete measurement of
every nucleotide component of DNA in a cell
(whole genome sequencing). Genetic biomarkers
are already being used to guide medical decision
making in determining cancer treatment 1 and
there is intense interest and activity in using
and identifying genetic biomarkers to improve
clinical trials. 2 RNA is another key component
of the cell and was one of the
fluids such as urine
and blood. All of these life science techniques
involve the acquisition and subsequent analysis
of large variable data sets. Multivariate statistical
analysis is an essential tool for the analysis and
interpretation of this data. This chapter focuses
on the primary methods of modern multivariate
analysis: principal component analysis (PCA)
and partial least squares (PLS). These essential
methods are
first to allow for
rst
explained using simple
 
Search WWH ::




Custom Search