Biology Reference
In-Depth Information
disease. This will enable the five holy grails of blood
disease diagnosis: (1) presymptomatic diagnosis; (2)
stratification of a disease into its different subtypes; (3)
assessment of the progression of the disease; (4)
following patient response to therapy; and (5) identi-
fying recurrences. We are now applying this strategy to
identify human organ-specific blood biomarkers for
several cancer types. In addition to blood proteins as
tumor biomarkers, circulating DNAs, mRNAs, and
microRNAs, as well as circulating tumor cells have also
been studied which can serve as surrogate disease
biomarkers and for monitoring cancer
50
100 base pair reads) which have enabled the rapid
sequencing of human genomes, the direct sequence
analyses of transcriptomes and miRNAomes, as well as
the analyses of some epigenetic features such as
methylation. DNA arrays remain useful for looking at
genetic variation and at
e
the quantification of RNA
populations
e
although it is clear that in time DNA sequencing will
replace these DNA array analyses. There is now
emerging a third generation of DNA sequencing
instruments that employ nanopores for threading single
DNA molecules through the pores to enable the elec-
tronic analysis of single-stranded DNA molecules
[44]
.
These new techniques have the potential for extremely
long sequence reads (50
(transcriptomes
and miRNAomes)
recurrence
32]
. The organ-specific blood fingerprints are
powerful aids to disease diagnostics, assessing drug
toxicities, validating the orthologies between human
disease and animal models of that disease, assessing
multiorgan responses to diseases (beginning to define
the organ/organ communicating networks (
Figure 23.4
)
and, as we shall discuss later, assessing longitudinally
across time the wellness of individual patients).
4.
The systems approach to disease mandates the need to
develop new or emerging technologies that can explore
new dimensions of patient data space as reflected in part
by the dynamics of the network of networks. These
technologies include new approaches to genomics,
proteomics, metabolomics, interactomics, cellomics,
organomics, in vitro and in vivo imaging, and other
high-throughput phenotypic measurements
[33
[28
e
100 kb or more) and the
ability for such extensive parallelization of the
sequencing runs (through threading many DNA mole-
cules simultaneously through many nanopores) that one
may imagine doing a sequence run for a complete
human genome sequence in a fraction of an hour, rather
than the day to week or more that is required by current
NGS instruments.
NGS has enabled striking new strategies for gener-
ating data (e.g., RNAseq
e
the quantification of
e
complete
the
sequence analyses of all (most) of the exons in a genome
and family genome sequencing
transcriptomes,
exon sequencing
e
determining the
complete human genome sequences of all the members
of a family).
Complete family genome sequencing integrates
genetics with genomics and in doing so raises fascinating
possibilities for delineating diverse chromosomal
features. For example, in the sequence analysis of a family
of four where the mother and father were healthy and the
two children each had two recessive genetic diseases, we
had hoped to identify a modest number of gene candidates
to explain the two genetic diseases (
Figure 23.8
)
[33]
.To
our surprise, family genome sequences enabled far more
in theway of family analysis. First, wewere able to correct
about 70% of the DNA sequencing errors merely by
a consideration of the principles of Mendelian genetics.
Second, we were able to identify rare variants merely by
asking whether two or more members of the family
exhibited the variant (thus eliminating the possibility of
DNA sequencing errors). Third, we were able to deter-
mine the sites of chromosomal recombination, and
accordingly could determine complete chromosomal
haplotypes for eachmember of the family. This turned out
to be important, as it reduced the chromosomal space in
which disease genes might reside (by asking which
haplotype regions were shared by the diseased children
compared to their healthy parents). In a family of four the
two affected genes must reside within a defined quarter of
e
38]
.
Microfluidic and nanotechnology approaches are
moving many of these assays towards further minia-
turization, parallelization, automation and integration
of complex chemical procedures
[39
e
41]
. The areas of
in vitro imaging and high-throughput phenotypics
assays are going to contribute enormously to expanding
the data repertoires of a systems approach to disease.
But these new technologies must be driven by the real
needs of biology or medicine. The outcome is an
exponentially increasing ability to generate enormous
amounts of digitalized personal data
e
that
necessitates a mandate to translate these data into
knowledge. Let us briefly consider several areas in
which emerging technologies are or will
big data
e
e
transform
a systems approach to medicine.
DNA sequencing. Automated DNA sequencing was the
cornerstone technology for sequencing the genome
[42]
. Since the initial completion of the human genome
sequence
[43]
there has emerged a series of 'next-
generation sequencing (NGS)' technologies that have
exponentially increased the throughput of DNA
sequencing while bringing down the costs dramatically
(through parallelization and miniaturization of the
process). NGS generates short sequence reads (e.g.,
