Biology Reference
In-Depth Information
FIGURE 2
(A) Overlay of typical GC/TOFMS chromatograms of urine samples and (B) OPLS-DA scores plots obtained
from the urine analysis of bladder cancer (BC) and non-BC subjects (H) subjects. (C) Validation plot obtained from 100
permutation tests and (D) receiver operating characteristic (ROC) calculated using cross validated Y-predicted values of
PLS-DA model
(Adapted from reference #33 with permission from ACS publication.)
enables the detection of temporal progression of
toxicity in terms of its onset, progression, remis-
sion, and recovery.
16,17
continuously as a single batch. For clinical meta-
bolic pro
ling of the general human population,
it is essential to account for substantial diversity
between subjects due to uncontrolled factors
(e.g., diet, lifestyle, and gender). In recent years,
there has been a trend of performing large-scale
epidemiological studies to provide the statistical
power to detect the subtle differences between
subjects and reduce false discoveries of bio-
markers.
19
Metabonomic studies that monitor
the temporal dynamics of metabolic perturba-
tion are also larger in scale due to the number
of time points collected per subject. Such large-
scale studies are made possible by technolog-
ical advancement and maturity of the analytical
platform. However,
STRATEGIES TO ADDRESS
LARGE-SCALE METABONOMIC
INVESTIGATIONS
Most GC/MS-based metabonomic investiga-
tions performed to date have relatively small
sample size, due to the controlled nature of the
experiments in which treatment is the only vari-
able, treatment effect is large, and there is rela-
tively little biological variation. If the number
of samples is small, all samples can be analyzed
strategies
that address
Search WWH ::
Custom Search