Content uploaded by Peter D Paré
Author content
All content in this area was uploaded by Peter D Paré on Nov 26, 2015
Content may be subject to copyright.
Comment
www.thelancet.com/respiratory Published online September 22, 2015 http://dx.doi.org/10.1016/S2213-2600(15)00362-8
1
Towards an integrative genomics of lung function
Genetic variation is an important determinant of lung
function and risk of chronic obstructive pulmonary
disease (COPD). Genome-wide association studies
(GWAS), fi rst described a decade ago, have become the
standard method t o study genetic factors infl uencing
these and other complex (non-mendelian) traits and
diseases. GWAS can be used to identify novel biological
pathways for human disease and ultimately inform
diagnostics and therapeutics.
1
However, they have
several limitations. One is a lack of power. Although
GWAS in most diseases have identifi ed at least a few
genetic loci, important insights can only be gained
when patterns among many loci are recognised.
2
To
detect such patterns, however, large sample sizes are
needed, and these can be diffi cult to achieve. Another
major limitation is that GWAS generally implicate
regions of the genome rather than specifi c genes or
pathophysiological mechanisms.
A study by Ma’en Obeidat and colleagues,
3
published
in The Lancet Respiratory Medicine, helps to address
these issues. To better understand the genetic and
genomic mechanisms of airway obstruction, they
expand on recently published work in a COPD cohort
4
by overlaying the largest genome-wide association
for lung tissue gene expression (also known as an
expression quantitative trait locus, or eQTL, study)
with the largest GWAS for spirometry, together
with publicly available data on blood expression
and transcriptional profi les produced by bioactive
compounds, among others. Of the previously
reported genome-wide associations between genetic
variants and lung function, about half seem to aff ect
the expression of at least one gene in lung tissue,
providing some initial hints as to how these genetic
variants might aff ect lung function. For example, on
chromosome 4q24 the genetic variant most strongly
associated with reduced lung function is closest to
GSTCD, but aff ects NPNT (and not GSTCD) expression in
lung tissue and INTS12 in blood. The researchers then
take their analysis a step further, by adding spirometry
association results that did not reach genome-wide
signifi cance but overlapped with lung eQTLs. They fi nd
that this larger set of genes seems to be enriched for
specifi c regulatory elements, as well as infl ammatory,
immune response, and developmental pathways,
and that these enrichments might be tissue-specifi c.
They also identify evidence of pleiotropy and use the
Connectivity Map (Cmap) to identify candidates for
drug repositioning. A broad range of other analyses
were done, many of which are of individual interest,
such as the identifi cation of an interaction between
HHIP and PTCH1 eQTLs.
Most causal genetic variants are likely to be
regulatory and signifi cantly enriched for eQTLs.
5
Obeidat and colleagues make important progress
by demonstrating tissue-specifi c eQTL enrichment
in their data. It should be noted, however, that the
overlap between GWAS variants and eQTLs is far from
complete. A previous study in blood eQTLs found
that only 16% of GWAS associations were eQTLs;
6
in
another study of nine tissues, only 6% showed strong
overlap with the “best” eQTL in at least one tissue.
7
Importantly, in the fi rst example,
6
11% of frequency-
matched non-GWAS variants were also blood eQTLs.
Although the defi nition of overlap depends on
defi nitions, statistical methods, and sample size, these
data overall suggest that not all causal lung function
variants are eQTLs, and lung function variants that
are eQTLs require further confi rmation. This lack
of accuracy might help to explain counter-intuitive
fi ndings in Obeidat and colleagues study,
3
such as
enrichment in cell types such as colon smooth muscle,
and high discordance of expected causal direction of
eff ect.
This study has other limitations, many of which have
been identifi ed by the researchers. For example, their
analysis included mostly healthy participants, and did
not consistently adjust for smoking intensity (pack-
years); specifi c relevance of these results to COPD and
smoking susceptibility is not known. Expression was
measured in whole lung tissue, consisting of many
diff erent cell types; similarly, the Cmap database, which
was used to fi nd drug repurposing candidates, uses a
small set of cancer cell lines. The integrative analysis
method used a simple overlap at a predefi ned, fi xed
threshold; more sophisticated statistical methods to
integrate GWAS and eQTLs are available.
Despite these limitations, Obeidat and colleagues’
work is a valuable contribution. eQTLs have been
instrumental in identifying causal mechanisms in GWAS.
Lancet Respir Med 2015
Published Online
September 22, 2015
http://dx.doi.org/10.1016/
S2213-2600(15)00362-8
See
Online/Articles
http://dx.doi.org/10.1016/
S2213-2600(15)00380-X
Alfred Pasieka/Science Photo Library
For the Connectivity Map see
www.broadinstitute.org/cmap/
Comment
2
www.thelancet.com/respiratory Published online September 22, 2015 http://dx.doi.org/10.1016/S2213-2600(15)00362-8
In asthma, ORMDL3 and later GSDMB and ZPBP2 were
identifi ed by expression as potential eff ector genes;
8,9
in obesity, eQTLs provided important evidence for a
pathway acting through IRX3 and not FTO, as originally
postulated.
10
The genetic mechanisms underlying impaired lung
function and COPD are undoubtedly complex. Rapid
advances in technology allow both detailed and broad
examination at a level that is unprecedented. Future
studies will likely incorporate whole genome sequencing
with RNA sequencing of specifi c cell types and individual
cells, together with epigenetic profi ling, metabolomics,
and proteomics. These studies will use integrative
statistical methods to identify complex relations within
and between data types and disease subtypes. Large-
scale integrative genomics in respiratory disease is still
in its infancy. The work of Obeidat and colleagues is an
important fi rst step.
Michael H Cho
Channing Division of Network Medicine and Division of
Pulmonary and Critical Care Medicine, Brigham and Women’s
Hospital, Boston, MA 02115, USA
remhc@channing.harvard.edu
I am supported by grants from the National Institute of Health (K08 HL097029,
R01 HL113264) and the Alpha-1 Foundation. I declare no other competing
interests. The content is solely my responsibility and does not necessarily
represent the offi cial views of the National Heart, Lung, and Blood Institute or
the National Institutes of Health. I thank Edwin K Silverman, Peter J Castaldi, and
Craig P Hersh for their helpful comments.
1 Visscher PM, Brown MA, McCarthy MI, et al. Five years of GWAS discovery.
Am J Hum Genet 2012; 90: 7–24.
2 Ripke S, Neale BM, Corvin A, et al. Biological insights from
108 schizophrenia-associated genetic loci. Nature 2014; 511: 421–27.
3 Obeidat M, Hao K, Bossé Y, et al. Molecular mechanisms underlying
variations in lung function: a systems genetics analysis. Lancet Respir Med
2015; published online Sept 22. http://dx.doi.org/10.1016/S2213-
2600(15)00380-X.
4 Castaldi PJ, Cho MH, Zhou X, et al. Genetic control of gene expression at
novel and established chronic obstructive pulmonary disease loci.
Hum Mol Genet 2015; 24: 1200–10.
5 Nicolae DL, Gamazon E, Zhang W, et al. Trait-associated SNPs are more
likely to be eQTLs: annotation to enhance discovery from GWAS.
PLoS Genet 2010; 6: e1000888.
6 Lappalainen T, Sammeth M, Friedländer MR, et al. Transcriptome and
genome sequencing uncovers functional variation in humans. Nature 2013;
501: 506–11.
7 Ardlie KG, Deluca DS, Segre AV, et al. The Genotype-Tissue Expression
(GTEx) pilot analysis: multitissue gene regulation in humans. Science 2015;
348: 648–60.
8 Verlaan DJ, Berlivet S, Hunninghake GM, et al. Allele-specifi c chromatin
remodeling in the ZPBP2/GSDMB/ORMDL3 locus associated with the risk
of asthma and autoimmune disease. Am J Hum Genet 2009; 85: 377–93.
9 Moff att MF, Kabesch M, Liang L, et al. Genetic variants regulating ORMDL3
expression contribute to the risk of childhood asthma. Nature 2007;
448: 470–73.
10 Smemo S, Tena JJ, Kim K-H, et al. Obesity-associated variants within FTO
form long-range functional connections with IRX3. Nature 2014;
507: 371–75.
