Ience Rapid Whole-Genome Sequencing for Genetic Disease Diagnosis in
Translational Neonatal Intensive Care Units
Medicine
AAAS
Carol Jean Saunders et at
Sci Trans! Med 4, 154ra135 (2012);
DOI: 10.1126/scitranslmed.3004041
Editor's Summary
Speed Heals
The waiting might not be the hardest part for families receiving a diagnosis in neonatal intensive care units
(NICUs), but it can be destructive nonetheless. While they wait on pins and needles for their newborn baby's
diagnosis, parents anguish, nurture false hope, wrestle with feelings of guilt—and all the while, treatment and
counseling are delayed. Now, Saunders et at describe a method that uses whole-genome sequencing (WGS) to
achieve a differential diagnosis of genetic disorders in 50 hours rather than the 4 to 6 weeks.
Many of the -3,500 genetic diseases of known cause manifest symptoms during the first 28 days of life, but full
clinical symptoms might not be evident in newborns. Genetic screens performed on newborns are rapid, but are
designed to unearth only a few genetic disorders, and serial gene sequencing is too slow to be clinically useful.
Together, these complicating factors lead to the administration of treatments based on nonspecific or obscure
symptoms, which can be unhelpful or dangerous. Often, either death or release from the hospital occurs before the 0
diagnosis is made.
The new WGS protocol cuts analysis time by using automated bioinformatic analysis. Using their newly 2
developed protocol, the authors performed retrospective 50-hour WGS to confirm, in two children, known molecular
diagnoses that had been made using other methods. Next, prospective WGS revealed a molecular diagnosis of a u-
BRAT1-related syndrome in one newborn; identified the causative mutation in a baby with epidermolysis bullosa:
ruled out the presence of defects in candidate genes in a third infants; and, in a pedigree, pinpointed BCL9L as a new co
recessive gene (HTX6) that gives rise to visceral heterotaxy —the abnormal arrangement of organs in the chest and $6
abdominal cavities. WGS of parents or affected siblings helped to speed up the identification of disease genes in the icf
prospective cases. These findings strengthen the notion that WGS can shorten the differential diagnosis process and E
quicken to move toward targeted treatment and genetic and prognostic counseling. The authors note that the speed c
and cost of WGS continues to rise and fall, respectively. However, fast WGS is clinically useful when coupled with fast
and affordable methods of analysis. -a
E
E
A complete electronic version of this article and other services, including high-resolution figures, -§
can be found at:
http://stm.sciencemag.org/content/4/154/154ra135.full.html a
Supplementary Material can be found in the online version of this article at:
http://stm.sciencemag.org/content/suppl/2012/10/01/4.154.154ra135.DC1.html
Related Resources for this article can be found online at:
http://stm.sciencemag.org/content/scitransmed/5/194/194cm5.full.html
http://stm.sciencemag.org/content/scitransmed/5/194/194ed10.full.html
http://stm.sciencemag.org/content/scitransmed/5/198/198ed12.full.html
http://www.sciencemag.org/content/sci/342/6155/197.full.html
Information about obtaining reprints of this article or about obtaining permission to reproduce this
article in whole or in part can be found at:
http://www.sciencemag.org/about/permissions.dtl
Science Translational Medicine (print ISSN 1946-6234; online ISSN 1946-6242) is published weekly, except the
last week in December. by the American Association for the Advancement of Science, 1200 New York Avenue
NW, Washington, DC 20005. Copyright 2012 by the American Association for the Advancement of Science; all
rights reserved. The title Science Translational Medicine is a registered trademark of AAAS.
EFTA00315100
RESEARCH ARTICLE
DIAGNOSTICS
Rapid Whole-Genome Sequencing for Genetic Disease
Diagnosis in Neonatal Intensive Care Units
Carol Jean Saunders,1,2,3,4,5* Neil Andrew MiIler,l '2'4* Sarah Elizabeth Soden,"4*
Darrell Lee Dinwiddie,"31"* Aaron Noll,' Noor Abu Alnadi,4 Nevene Andraws,3
Melanie LeAnn Patterson," Lisa Ann Krivohlavek," Joel Ferns,' Sean Humphray,' Peter Saffrey,'
Zoya Kingsbury,' Jacqueline Claire Weir,' Jason Betley," Russell James Grocock,'
Elliott Harrison Margulies,' Emily Gwendolyn Farrow,' Michael Artman,2'4 Nicole Pauline Safina,"
Joshua Erin Petrikin," Kevin Peter Hall,' Stephen Francis Kingsmore'•2,3,4,st
Monogenic diseases are frequent causes of neonatal morbidity and mortality, and disease presentations are often
undifferentiated at birth. More than 3500 monogenic diseases have been characterized, but clinical testing is avail-
able for only some of them and many feature clinical and genetic heterogeneity. Hence, an immense unmet need
exists for improved molecular diagnosis in infants. Because disease progression is extremely rapid, albeit hetero-
geneous, in newborns, molecular diagnoses must occur quickly to be relevant for clinical decision-making. We de- 0
scribe 50-hour differential diagnosis of genetic disorders by whole-genome sequencing (WGS) that features
automated bioinformatic analysis and is intended to be a prototype for use in neonatal intensive care units. Ret-
rospective 50-hour WGS identified known molecular diagnoses in two children. Prospective WGS disclosed potential
molecular diagnosis of a severe G22-related skin disease in one neonate; BRAT/-related lethal neonatal rigidity and
multifocal seizure syndrome in another infant identified Ba9L as a novel, recessive visceral heterotaxy gene (H7X6) in a a)
pedigree; and ruled out known candidate genes in one infant. Sequencing of parents or affected siblings expedited the
C
identification of disease genes in prospective cases. Thus, rapid WGS can potentially broaden and foreshorten differ- 0
ential diagnosis, resulting in fewer empirical treatments and faster progression to genetic and prognostic counseling. rn
til
INTRODUCTION rapid, identify only a few genetic disorders for which inexpensive tests E
Genomic medicine is a new, structured approach to disease diagnosis and cost-effective treatments exist (14, 15). Further complicating diag- 8 C
tb
and management that prominently features genome sequence infor- nosis is the fact that the full clinical phenotype may not be manifest •O
mation (1). Whole-genome sequencing (WGS) by next-generation in newborn infants (neonates), and genetic heterogeneity can be im- N
sequencing (NGS) technologies has the potential for simultaneous, mense. Thus, acutely ill neonates with genetic diseases are often dis- E
comprehensive, differential diagnostic testing of likely monogenic ill- charged or deceased before a diagnosis is made. As a result, NICU
E
nesses, which accelerates molecular diagnoses and minimizes the du- treatment of genetic diseases is usually empirical, may lack efficacy,
ration of empirical treatment and time to genetic counseling (2-7). may be inappropriate, or may cause adverse effects.
Indeed, in some cases, WGS or exome sequencing provides molecular NICUs are also suitable for early adoption of genomic medicine
diagnoses that could not have been ascertained by conventional single- because extraordinary interventional efforts are customary and inno-
gene sequencing approaches because of pleiotropic clinical presenta- vation is encouraged. Indeed, NICU treatment is among the most
tion or the lack of an appropriate molecular test (7-9). cost-effective of high-cost health care, and the long-term outcomes of
Neonatal intensive care units (NICUs) are especially suitable for most NICU subpopulations are excellent (16-18). In genetic diseases 8
early adoption of diagnostic WGS because many of the 3528 mono- for which treatments exist, rapid diagnosis is critical for timely delivery
genic diseases of known cause are present during the first 28 days of of interventions that lessen morbidity and mortality (14-17, 19, 20).
life (10). In the United States, more than 20% of infant deaths are For neonatal genetic diseases without effective therapeutic interven-
caused by congenital malformations, deformations, and chromosomal tions, of which there are many (21), timely diagnosis avoids futile inten-
abnormalities that cause genetic diseases (11-13). Although this pro- sive care and is critical for research to develop management guidelines
portion has remained unchanged for the past 20 years, the precise that optimize outcomes (22).In addition to influencing treatment, neo-
prevalence of monogenic diseases in NICUs is poorly understood be- natal diagnosis of genetic disorders and genetic counseling can spare
cause ascertainment rates are low. Serial gene sequencing is too slow parents diagnostic odysseys that instill inappropriate hope or perpetuate
to be clinically useful for NICU diagnosis. Newborn screens, while needless guilt
Two recent studies exemplify the diagnostic and therapeutic uses
of NGS in the context of childhood genetic diseases. WGS of fraternal
'Center for Pediatrocancenc Medidne Children's Mercy Hospital• Kansas City,M0 69103,
USA /Department of Pediatric,. Children's Mercy Hospital. Kansas City. MO 6910. IA&
twins concordant for 3,4-dihydroxyphenylalanine (dopa)-responsive
'Department of Pathology. Children's Mercy Hosptat Kansas City. M069103. USA °School dystonia revealed known mutations in the sepiapterin reductase
of Medicine. University of Missoun-Kansas Cny. Kansas City. MO 6410& USA. sUni.ersity of (SPR) gene (3). In contrast to other forms of dystonia, treatment with
Kansas Medical Center. KansisCny. KS66160, USA'1Ilumina Inc,Chesterfad Research Park 5-hydroxytryptamine and serotonin reuptake inhibitors is beneficial in
Little Chesterford. CBIO I XL Essex UK
'These authors contributed equally to this work patients with SPR defects. Application of this therapy in appropriate
tic, whom correspondence should be addressed. E-mail stkingsrnore@cmhedu cases resulted in clinical improvement. Likewise, extensive testing
wiwSdenceTranslatIonalMedlamorg 3 October 2012 Vol 4 Issue 154 154ral 3S 1
EFTA00315101
RESEARCH ARTICLE
failed to provide a molecular diagnosis for a child with fulminant pan- per patient (maximum, 430; minimum, 5). Thus, SSAGA displayed
colitis (extensive inflammation of the colon) (8), in whom standard sufficient sensitivity for the initial selection of known, recessive candi-
treatments for presumed Crohn's disease—an inflammatory bowel date genes in children with specific clinical presentations.
disease—were ineffective. NGS of the patient's exome, together with
confirmatory studies, revealed X-linked inhibitor of apoptosis (XIAP) Rapid WGS
deficiency. The treating physicians had not entertained this diagnosis To assess our ability to recapitulate known results, we performed rapid
because X1AP mutations had not previously been associated with co- WGS retrospectively on DNA samples from two infants with molec-
litis Hemopoietic progenitor cell transplant was performed, as indi- ular diagnoses that had previously been identified by clinical testing.
cated for XIAP deficiency, with complete resolution of colitis. last, Then, to assess the potential diagnostic use of rapid WGS, we prospec-
for —3700 genetic illnesses for which a molecular basis has not yet tively performed WGS in five undiagnosed newborns with clinical
been established (10), WGS can suggest candidate genes for functional presentations that strongly suggested a genetic disorder as well as their
and inheritance-based confirmatory research (23). siblings.
The current cost of research-grade WGS is $7666 (24)-which is Automation of the five main components of WGS as well as
similar to the current cost of commercial diagnostic dideoxy sequencing bioinformatics-based gene-variant characterization and clinical inter-
of two or three disease genes. Within the context of the average cost per pretation, all in an integrated workflow, made possible —50-hour time
day and per stay in a NICU in the United States (13), WGS in care- to differential molecular diagnosis of genetic disorders (Fig. 1).
fully selected cases is acceptable and even potentially cost-saving (3-7). Specifically, sample preparation for WGS was shortened from 16 to
However, the turnaround time for interpreted WGS results, such as 4.5 hours, while a physician simultaneously entered into SSAGA clin-
that of dideoxy sequencing, is too slow to be of practical use for NICU ical terms that described the neonates' illnesses (fig. SI). For each sample,
diagnoses or clinical guidance (typically -4 to 6 weeks) (2-4). Here, we rapid WGS [2 x 100 base pair (bp) reads, including on-board cluster
report a system that permits WGS and bioinfonnatic analysis (largely generation and paired-end sequencing] was performed in a single run as as
automated) of suspected genetic disorders within 50 hours, a time frame on the alumina HiSeq 2500 and took —26 hours. Base calling, genomic 2
that appears to be promising for emergency use in level 3 NICUs. sequence alignment, and gene variant calling took —15 hours. The HiSeq .o
2500 runs yielded 121 to 139 gigabases (GB) of aligned sequences (34- ts.
to 4I-fold aligned genome coverage; 'fable I). Eighty-eight to 91% of 5
RESULTS bases had >99.9% likelihood of being correct (quality score >30, using ,c2)
Illumina software equivalent to Phred) (31, 32). We detected 4.00 ± of
Symptom- and sign-assisted genome analysis (SSAGA) is a new din- 0.20 million nudeotides that differed from the reference genome se-
icopathological correlation tool that maps the clinical features of 591 quence (variants) (mean ± SD) in nine samples, one from each of nine E
well-established, recessive genetic diseases with pediatric presentations infants (Table 1). C
(table SI) to corresponding phenotypes and genes known to cause the N
symptoms (2, 10). SSAGA was developed for comprehensive auto- Analytical metrics
mated performance of the following two tasks: (i) WGS analyses re- In three samples, genome variants identified by 50-hour WGS were E
stricted to a superset of gene-associated regions relevant to clinical compared with those identified by deep targeted sequencing of either
E
presentations, in accord with published guidelines for genetic testing exons and 20 intron-exon boundary nucleotides of a panel of 525 re-
in children (25-28), and (ii) prioritization of clinical information to cessive disease genes [Children's Mercy Hospital Diagnostic panel
assist in the interpretation of WGS results. SSAGA has a menu of 1 (CMH-Dxl)] or the exome (Table 2). CMH-Dx1 comprised 8813 -0
227 clinical terms arranged in nine symptom categories (fig. SI). Stan- exonic and intronic targets, totaling 2.1 million nucleotides (table SI) cu
dardized clinical terms (29) have been mapped to 591 genetic diseases (2, 33). The exome and CMH-Dx1 methods, which used Illumina
on the basis of authoritative databases (10, 30) and expert physician TruSeq enrichment and HiSeq 2000 sequencing, took —19 days. In
reviews. Each disease gene is represented by an average of 8 terms and contrast, rapid WGS did not use target enrichment, was performed 8
at most 11 terms (minimum, I term, 15 disease genes; maximum, 11 with the HiSeq 2500 instrument, and took —50 hours. Samples
terms, 3 disease genes). CMH064, UDT002, and UDTI73 were sequenced using these three
To validate the feasibility of automated matching of clinical terms methods, and variants were detected with a single alignment method
to diseases and genes, we entered retrospectively the presenting fea- [the Genomic Short-read Nucleotide Alignment Program (GSNAP)]
tures of 533 children who have received a molecular diagnosis at (34) and variant caller [the Genome Analysis Tool Kit (GATK)] (35).
our institution [Children's Mercy Hospital (CMH), Kansas City, MO] Rapid WGS detected —96% of the variants identified by a target en-
within the last 10 years into SSAGA. Sensitivity was 99.3% (529), as richment method and —99.5% of the variants identified by both
determined by correct disease and affected gene nominations. Failures methods had identical genotypes (Table 2), indicating that rapid
induded a patient with glucose-6-phosphate dehydrogenase deficiency WGS is highly concordant with established clinical sequencing
who presented with muscle weakness [which is not a feature men- methods (33). In contrast, analysis of the rapid WGS data set from
tioned in authoritative databases (10, 30)]; a patient with Janus kinase sample CMH064 with three different alignment and variant detection
3 mutations who had the term "respiratory infection" in his medical methods [GSNAP/GATK, the alumina CASAVA alignment tool, and
records rather than "increased susceptibility of infections," which is the Burrows-Wheeler Alignment (BWA) tool] revealed surprising dif-
the description in authoritative databases; and a patient with cystic ferences between the variants detected. Only —80% of the variants de-
fibrosis who had the term "recurrent infections" in his medical records tected using GATK/GSNAP or BWA were also detected with CASAVA
rather than "respiratory infections," which is the description in au- (Table 2 and table S2) (36-41). This suggests that additional studies will
thoritative databases. SSAGA nominated an average of 194 genes be needed to define optimal alignment methods for dinical sequencing.
wwwSdenceTranslatIonalMediclnenrg 3 October 2012 Vol 4 Issue 154 154ra135 2
EFTA00315102
RESEARCH ARTICLE
Nevertheless, there was good concordance between the genotypes of
variants detected by rapid WGS (using the HiSeq 2500 and CASAVA)
and targeted sequencing (using exome enrichment, the HiSeq 2000, and
GATKIGSNAP)-99.5% (UDT002), 99.9% (UDT173), and 99.7%
(CMH064) (Table 2)—further indicating that rapid WGS is highly con-
cordant with an established genotyping method (33). In subsequent
Obtain consent and blood sample studies, the rapid WGS technique used CASAVA for alignment and
variant detection.
Genomic variants were characterized with respect to functional
consequence and zygosity with a new software pipeline [Rapid
Understanding of Nucleotide variant Effect Software (RUNES), fig.
S21 that analyzed each sample in 2.5 hours. Samples contained a mean
of 4.0D ± 0.20 million (SD) genomic variants, of which a mean of 1.87 ±
0M9 million (SD) were se-striated with protein-encoding genes (Table 1).
Prepare sequencing library Less than I% of these variants (mean, 10,848 ± 523 SD) were also of a
functional class that could potentially be disease causative (Fable 1)
Enter clinical findings into SSAGA (25-27). Of these, —14% (mean, 1530 ± 518 SD) had an allele fiequen-
cy that was sufficiently low to be a candidate for being causative in an
uncommon disease (<1% allele frequency in 836 individuals sequenced d
at CMH) (42). Last, of these, —71% (mean, 1083 ± 240 SD) were also of
S
a functional class that was likely to be disease causative [American i•
College of Medical Genetics (ACMG) categories Ito 31 (Table 1). This 2
set of variants was evaluated for disease causality in each patient, with
priority given to variants within the candidate genes that had been Li-
HiSeq 2500 2 x 100 bp sequencing nominated by an individual patient presentation.
Es
Retrospective analyses °
Patient UDT002 was a male who presented at 13 months of age with na
hypotonia, developmental regression. Brain magnetic resonance imag- 5
ing (MRI) showed diffuse white matter changes suggesting leukodys-
trophy. Three hundred fifty-two disease genes were nominated by .2 0
one of the three clinical terms hypotonia, developmental regression, or 01
CASAVA base calling leukodystrophy, 150 Aise-ase genes were nominated by two terms; and
9 disease genes were nominated by all three terms (table S3). Only E
RUNES variant annotation 16 known pathogenic variants had allele frequencies in dbSNP and the 2
CMH cumulative database that were consistent with uncommon dis-
ease mutations. Of these, only two variants mapped to the nine can-
didate genes; the variants were both compound heterozygous (verified cog,
by parental testing) substitution mutations in the gene that encodes the t
a subunit of the lysosomal enzyme hexosaminidase A [HEXA Chr
15:72,641,417c>C (gene symbol, chromosome number, chromosome
coordinate, reference nucleotide > variant nucleotide), c.986+3A>G
SSAGA-delimited variant analysis (transcript coordinate, reference nucleotide, variant nucleotide), and
and interpretation Chr15:72,640,388C>T, c.1073+1G>A1. The c.986+3A>G alters a 5'
exon-flanking nucleotide and is a known mutation that causes Tay-Sachs
disease ('15D), a debilitating lysosomal storage disorder [Online
Mendelian Inheritance in Man (OMIM) number 2728001. The variant
had not previously been observed in our database of 651 individuals or
dbSNP, which is relevant because mutation databases are contaminated
with some common polymorphisms, and these can be distinguished from
true mutations on the basis of allele frequency (33). The c1073+1G>A
Verbal interim report of diagnosis variant is a known l'SD mutation that affects an exonic splice donor site
pending CLIA confirmation (dbSNP r576173977). The variant has been observed only once before
in our database of 414 samples, which is consistent with an allele frequen-
cy of a causative mutation in an orphan genetic disease. Thus, the known
Fig. 1. STAT-Sea. Summary of the steps and timing of STAT-Seq, result- diagnosis of 'Est) was confirmed in patient UDT002 by rapid WGS.
ing in an interval of 50 hours between consent and delivery of a pre- Patient UDT173 was a male who presented at 5 months of age with
liminary, verbal diagnosis. t, hours. developmental regression, hypotonia, and seizures. Brain MRI showed
www.5cienceTranslationalMedlciae.org 3 October 2012 Vol 4 Issue 154 154ral 35 3
EFTA00315103
RESEARCH ARTICLE
dysmyelination, hair shaft analysis revealed pill ford (kinky hair), and (B1nrlec SUbstitution Matrix) scores. The known diagnosis of Menkes
serum copper and ceruloplasmin were low. On the basis of this clinical disease (OMIM number 309400) was recapitulated. As a further assess-
presentation, 276 disease genes matched one of these clinical terms and ment of the reliability of variant detection of rapid WGS, samples
3 matched three terms (table S4). 'There were no previously reported UDT002 and UDTI73 were aligned to the reference genome with three
disease-causing variants in these 276 genes. However, five of the candi- different alignment methods. The causative variants were recovered
date genes contained either variants of a type that is expected to be with each method.
disease-causing based on their predicted functional consequence or
missense variants of unknown significance (VUS). One of these var- Prospective analyses
iants was in a gene that matched all three clinical terms and was a Mutations in 35 genes can cause generalized, erosive dermatitis of the
hemizygous substitution mutation in the gene that encodes the a poly- type found in CMH064 (table S5). The severe phenotype, negative
peptide of copper-transporting adenosine triphosphatase (ATP7A Chr family history, and absence of consanguinity suggested dominant de
)C77,271,307C>T, c.2555C>T, i, aberrant forms of which are novo or recessive inheritance. No known pathogenic mutations were
known to cause Menkes disease, a copper-transport disorder. This identified in the candidate genes that had low allele frequencies in the
variant—new to our database and dbSNP—specified a nonconserva- CMH cumulative genome and exome sequence database and similar
tive substitution in an amino acid that was highly conserved across public datakcec Average coverage of the genomic regions corresponding
species and had deleterious SIFT (Sons Intolerant From Tolerant sub- to the candidate genes was 38.9-fold, and 98.4% of candidate gene
stitutions), PolyPhen2 (Polymorphism Phenotyping), and BLOSUM nucleotides had >I6x high-quality coverage (sufficient to rule out a n
Table 1. Sequencing, alignment, and variant statistics of nine samples analyzed by rapid WGS. ACMG category 1 to 4 variants are a subset of gene- (5
associated variants. L'"
ets
ACMG ACMG n2
High- ACMG categories categories Candidate or
Run Mitochondria Nuclear Gene- nor
Sequence quality categories 1 to 4 1 to 3 Candidate gene Candidate c
Sampletime genome )genome
°
(hours) variants variants variants
(%) variants frequency frequency 1 variants
<1% <1% 0)
to
UDT002 255 133 91 33 4,014,761 1,888,650 10,733 1,989 1,330 352 (9) 2 0 E
UDT173 255 139 89 40 3,977,062 1,859,095 10,501 2,190 1,296 347 (3) 0 I C
CD
CMH064 26.6 121 88 41 3,985,929 1,869,515 10,701 1.884 1,348 35 0 2
to
CMH076 25.7 134 88 34 4,498,146 2,098,886 11,891 2,552 1,351 89 0
CMH172 265 113 91 39 3,759,165 1,749,868 10,135 1,456 982 174 0 N
CMH184 265 137 90 37 3,921,135 1,840,738 10,883 833 12 0 0 E
1,168
2
CMH185 40 117 93 37 3,922,736 1,831,997 10,810 1,164 840 14 0 0
CMH186 255 113 93 37 3,933,062 1,827,499 10,713 1,202 868 14 §
CMH2O2 40 116 93 39 3,947,053 1,849,647 10,805 1,283 901 C
C
00
Table 2. Variants and genotypes. Comparisons of variants and genotypes for 523 genes and Merl 2000 sequencing. Average coverage of target nucleo-
obtained In three samples using three target enrichment methods, two se- tides indicates the average aligned sequence depth over the corresponding
quencing methods, and two alignment methods. The SO hour WGS (STAT-Seci) target panel. For WGS, the target is the genome; for come sequencing, the
was not enriched and used HiSeq 2500 sequencing. CMHDx1 was erviched target is the exome and for CMH-Corl, the targets are 523 genes.
Sample Target Sequencing Alignment Sequence Average coverage of Variants detected Genotypes Identical to
enrichment method method (GB) target nucleotides by rapid WGS both methods (%)
CMH064 Exorne HiSeq 2000 GATK/GSNAP 9.8 79 46,756 (96.0%) 99.4
None (WGS) HiSeq 2500 12.1 40
UDT173 CMH-Dxl HiSeq 2000 GATK/GSNAP 4.1 784 1539 (96.7%) 99.60
None (WGS) HiSeq 2500 13.9 46
UDT173 CMH-Dxl HiSeq 2000 GATK/GSNAP 4.1 784 1457 (83.0%) 99.9
None (WGS) HiSeq 2500 CASAVA 13.9 46
UDT002 CMH-Dxl HiSeq 2000 GATK/GSNAP 4.2 770 1341 (76.6%) 99.5
None (WGS) HiSeq 2500 CASAVA 13.3 44
www.ScienceTranslationalMedicine.org 3 October 2012 Vol 4 Issue 154 154ra135 4
EFTA00315104
RESEARCH ARTICLE
heterozygous variant; table S6). Five candidate genes had 100% nu- (46). Dideoxy sequencing confirmed the variant to be homozygous in
cleotides with >16-fold high-quality coverage and, thus, lacked a known CMH172 and heterozygous in both parents.
pathogenic mutation in an exon or within 20 nucleotides of the intron- Rapid WGS was performed simultaneously on proband CMHI84
exon boundaries. Eighteen of the candidate genes had >99% nucleotides (male), affected sibling (brother) CMHI85, and their healthy parents,
with >16-fold high-quality coverage, and 31 had >95% nucleotides with CMH186 and CMH2O2. Twelve genes have been associated with the
at least this level of coverage. Furthermore, while 26 of the candidate clinical features of the brothers (heterotaxy and congenital heart dis-
genes had pseudogenes, paralogs, and/or repeat segments (table S6) ease table S9). Co-occurrence in two siblings strongly suggested reces-
that could potentially result in misalignment and variant miscalls, only sive inheritance. No known disease-causing variants or homozygous/
0.03% of target nucleotides had poor alignment quality scores. compound heterozygous VUS with low allele frequencies were identi-
Among the 35 candidate genes nominated by the phenotype, two fied in these genes. A genome-wide search of novel, homozygous/
rare heterozygous VUS were detected in CMH064; however, dideoxy compound heterozygous, likely pathogenic VUS that were common
sequencing of both healthy parents exduded one, in the keratin 14 to the affected brothers and heterozygous in their parents yielded
gene, as a de novo mutation. The exomes of both parents were sub- two nonsynonymous variants in the B cell CLL/ghc -like gene
sequently sequenced, and variants were examined in the trio at length. (BCL9L, Chr 11:118,772,350G>A,c.2102G>A, and Chr
Three likely de novo mutations with excellent sequence coverage were 11:118,774,140G>A, c.554C>T, i. Evidence supporting the
identified in disease-causing genes. Of these, one was a candidate gene candidacy of BCL9L for heterotaxy and congenital heart disease is
for CMH064. It was an in-frame deletion of three nucleotides in GIB2, presented below.
(NM_004004), which encodes the connedn 26 protein. The variant, 0
c85_87de1, removes a highly conserved amino acid within 0
the first transmembrane helix (43). Dideoxy sequencing confirmed it DISCUSSION
to be a de novo mutation. Dominant, de novo 6,1B2 mutations have a's
been associated with severe neonatal lethal disorders of the skin, such Genomic medicine, empowered by WGS, has been heralded as trans- g
as keratitis-ichthyosis-deafness syndrome (KIDS), that involve the formational for medical practice (2, 4, 5, 47). Over the last several y
suprabasilar layers of the epidermis (OMIM number 148210) (44). The years, the cost of WGS has fallen markedly, potentially bringing it LI-
phenotype of CMH064 was atypical for KIDS, and functional studies within the realm of cost-effectiveness for high-intensity medical prac-
are in progress to determine causality definitively. tice, such as occurs in NICUs (3, 8, 23, 24). Furthermore, experience IF
Diagnoses suggested by the presentation in CMH076 were mito- has been gained with clinical use of WGS that has instructed initial ;
chondrial disorders, organic acidemia, or pyruvate carboxylase defi- guidelines for its use in molecular diagnosis of genetic disorders (9).
ciency. Together, 75 nuclear genes and the mitochondrial genome However, a major impediment to the implementation of practical ge-
cause these diseases (table S7). A negative family history suggested re- nomic medicine has been time to result
cessive inheritance that resulted from compound heterozygous or hemi- This limitation has always been a problem for diagnosis of genetic .0 a)
zygous variants or a heterozygous de novo dominant variant. Rapid disease& Time to result and cost have greatly constrained the use of 1.'1
WGS excluded known pathogenic mutations in the candidate genes. serial analysis of single-gene targets by dideoxy sequencing, Hitherto,
One novel heterozygous VUS was found. However, de novo occur- clinical use of WGS by NGS has also taken at least a month: Sample
rence of this variant was ruled out by exome sequencing of his healthy preparation has taken at least a day; clustering 5 hours; 2 x 100 nu- O
parents. No homozygous or compound heterozygous VUS with suit- cleotide sequencing 11 days; alignment, variant calling, and genotyping
ably low allele frequencies were identified in the known disease genes. 1 day, variant characterization a week and clinical interpretation at least
Potential novel candidates included 929 nuclear genes that encode a week Although exome sequencing lengthens sample preparation by 113
mitochondrial proteins but have not yet been associated with a genetic several days, it decreases computation time somewhat and is less costly. t
disease (45). Only one of these had a homozygous or compound het- For use in acute care, the turnaround time of molecular diagnosis, 3
erozygous VUS with an allele frequency in dbSNP and the CMH including analysis, must match that of medical decision-making, which
database that was sufficiently low to be a candidate for causality in ranges from Ito 3 days for most acute medical care. Herein, we de-
an uncommon inherited disease. Deep exome sequencing of both scribed proof of concept for 2-day genome analysis of acutely ill neo-
parents excluded this variant and did not disclose any further poten- nates with suspected genetic disorders.
tially causal variants.
A total of 174 genes are known to cause epilepsy of the type found Automating medicine
in CMH172 (table S8). A positive family history of neonatal epilepsy and Rapid WGS was made possible by two innovations. First, a widely
evidence of shared parental ancestry strongly suggested recessive inheri- used WGS platform has been modified to generate up to 140 GB of
tance. No known disease-causing variants or homozygous/compound sequence in less than 30 hours (HiSeq 2500): Sample preparation took
heterozygous VUS with low allele frequencies were identified in these 4.5 hours, and 2 x 100 bp genome sequencing took 25.5 hours (Fig. 1).
genes, which largely excluded them as causative in this patient A genome- The total "hands-on" time for technical staff was 5 hours. Modifica-
wide search of homozygous, likely pathogenic VUS that were novel in tions included a new flowcell design and faster imaging and chemistry.
the CMH database and dbSNP disclosed a frame-shifting insertion in Previously, NGS has either lacked sufficient sequence quantity, quality,
the BRCA)-associated protein required for ATM activation-I (BRAT), or read lengths for clinical use of WGS or been too slow for use in acute
Chr 7:2,583,573.2,583,574insATCITCTC,c453_454insATCITCTC, patient care. Rapid WGS generated -40-fold aligned genome coverage.
. A literature search yielded a very recent study of The sequence quality was very similar to that obtained with its predeces-
BRAT) mutations in two infants with lethal, multifocal seizures, hyper- sor (HiSeq 2000), as determined by quality scores and alignment rates
tonia, microcephaly, apnea, and bradycardia (OMINI number 614498) (48). Genotypes of nucleotide variants were >99.5% concordant with
wwwSdenceTranslatIonalMedicintorg 3 October 2012 Vol 4 Issue 154 154ral 35 5
EFTA00315105
RESEARCH ARTICLE
those of very deeply sequenced, partial exomes (33). The accuracy of the features. In male UDT173, a hemizygous (X-linked) VUS was identi-
latter has been extensively benchmariced and is >99.9% (33). fied in the single candidate gene matching all clinical features. The
Second, we automated much of the onerous characterization of ge- variant, a nonsynonymous nucleotide substitution, was predicted to
nome variation and facilitated interpretation by restricting and prior- be damaging. Rapid WGS also provided a definitive diagnosis in
itizing variants with respect to allele frequency (42), likelihood of a one of four infants enrolled prospectively. In CMH172, with refractory
functional consequence (25), and relevance to the prompting illness. epilepsy, rapid WGS disclosed a novel, homozygous frame-shifting
Thus, rapid WGS, as described herein, was designed for prompt dis- insertion in a single candidate gene (BRATI). BRATI mutations were
ease diagnosis rather than carrier testing or newborn screening. SSAGA very recently reported in two unrelated Amish infants who suffered
mapped the clinical features in ill neonates and children to disease lethal, multifocal seizures (46). A molecular diagnosis was reached
genes. Thereby, analysis was limited only to the parts of the genome within 1 hour of WGS data inspection in CMH172, even though a-
relevant to an individual patient's presentation, in accord with guide- tant reference databases [Human Gene Mutation Database (HGMD)
lines for genetic testing in children (25-28). This greatly decreased and OMIMI had not yet been updated with a BRATI disease associ-
the number of variants to be interpreted. In particular, SSAGA caused ation. The diagnosis was made clinically reportable by resequencing
most incidental (secondary) findings to be masked. In the setting of the patient and her parents. Had this diagnosis been obtained in real
acute care in the NICU, secondary findings are anticipated to impede time, it may have expedited the decision to reduce or withdraw
facile interpretation, reporting, and communication with physicians and support. The latter decision was made in the absence of a molecular
patients greatly (9, 49, 50). SSAGA also assisted in test ordering, diagnosis after 5 weeks of ventilatory support, testing, and unsuc- itr
permitting a broad selection of genes to be nominated for testing based cessful interventions to control seizures. Given high rates of NICU a
on entry of the patients' clinical features with easy-to-use pull-down bed occupancy, accelerated diagnosis by rapid WGS has the potential
menus. The version used herein contains —600 recessive and mitochon- to reduce the number of neonates who are turned away. The molec-
drial diseases and has a diagnostic sensitivity of 993% for those dis- ular diagnosis was also useful for genetic counseling of the infant's
orders. SSAGA is likely to be particularly useful in disorders that parents to share the information with other family members at risk 2
feature clinical or genetic heterogeneity or early manifestation of partial for carrying of this mutation. As suggested by recent guidelines (9), -,9
phenotypes because it maps features to a superset of genetic disorders. this case demonstrates the use of WGS for diagnostic testing when IL
SSAGA needs to be expanded to encompass dominant disorders and a genetic test for a specific gene of interest is not available.
to the full complement of genetic diseases that meet ACMG guidelines In four of five affected individuals, prospective, rapid WGS provided IF
for testing rare disorders (such as having been reported in at least two a definitive or likely molecular diagnosis in —50 hours. These cases dem- tel
unrelated families) (26). Although neonatal disease presentations are onstrated the use of WGS for diagnostic testing when a high degree of as
often incomplete, only one feature is needed to match a disease gene to genetic heterogeneity exists, as suggested by recent guidelines (9). Con- I
a presentation. In cases for which SSAGA-delimited genome analysis firmatory resequencing, which is necessary for return of results until rapid c
was negative, such as CMH064 and CMH076, a comprehensive second- WGS is compliant with Clinical laboratory Improvement Amendments c •—
oD
ary analysis was performed with limitation of variants solely to those (CLIA), took at least an additional 4 days. Until compliance has been
with acceptable allele frequencies (42) and likelihood of a functional established, we suggest preliminary verbal disclosure of molecular rt;
consequence (25). Nevertheless, secondary analysis was relatively facile, diagnoses to the neonatologist of record, followed by formal reporting E
yielding about 1000 variants per sample. upon performance of CLIA-conforming resequencing. Staged return of 2
RUNES performed many laborious steps involved in variant char- results of broad or complex screening tests, together with considered, a -
acterization, annotation, and conversion to HGVS (Human Genome pert interpretation and targeted quantification and confirmation, is likely .8
Variation Society) nomenclature in -2 hours. RUNES unified these in to be acceptable in intensive care. Precedents for rapid return of interim, 8
an automated report that contained nearly all of the information de- potentially actionable results include preliminary reporting of histo-
sirable for variant interpretation, together with a cumulative variant pathology, radiographic, and imaging studies and interim antibiotic selec- g
allele frequency and a composite ACMG categorization of variant tion based on Gram stains pending culture and sensitivity results.
pathogenicity (fig. S2). ACMG categorization is a particularly useful
standard for prioritization of the likelihood of variants being causal Disease gene sleuthing
(26). In particular, more than 75% of coding variants were of ACMG Because at least 3700 monogenic disease genes remain to be identified
category 4 (very unlikely to be pathogenic). Removal of such variants (10), WGS will often rule out known molecular diagnoses and suggest
allowed rapid interpretation of high-likelihood pathogenic variants in novel candidate disease genes (23, 51). Indeed, in another prospectively
relevant genes. The hands-on time for starting pipeline components enrolled family, WGS resulted in the identification of a novel candidate
and interpretation of known disease genes was, on average, less than disease gene, providing a likely molecular diagnosis. The proband
1 hour. Because genomic knowledge is currently limited to 1 to 2% of was the second affected child of healthy parents. Accurate genetic
physicians (physician scientists, medical geneticists, and molecular counseling regarding risk of recurrence had not been possible because
pathologists), variant characterization, interpretation, and clinical guid- the first affected child lacked a molecular diagnosis. We undertook
ance tools are greatly needed, as is training of medical geneticists and rapid WGS of the quartet simultaneously, allowing us to further limit
genetic counselors in their use. incidental variants by requiring recessive inheritance. Rapid WGS ruled
out 14 genes known to be associated with visceral heterotaxy and con-
Return of results genital heart disease (HTX). Among genes that had not been associated
In blinded, retrospective analyses of two patients, rapid WGS correctly with HTX, rapid WGS of the quartet narrowed the likely pathogenic
recapitulated known diagnoses. In child UDT002, two heterozygous, variants to two in the BCL9L gene BCL9L had not previously been as-
known mutations were identified in a gene that matched all clinical sociated with a human phenotype but is an excellent candidate gene for
www.ScienceTranslationalMediclne.org 3 October 2012 Vol 4 Issue 54 154ral 3S 6
EFTA00315106
RESEARCH ARTICLE
HTX based on its role in the Wingless (Writ) signaling pathway, which nucleotide in the genome (71). At 50x aligned coverage of the genome,
controls numerous developmental processes, including early embryonic WGS genotyped at least 95% of the reference genome with greater than
patterning, epithelial-mesenchymal interactions, and stem cell mainte- 99.95% accuracy, using methods very similar to those used in this study
nance (51, 52). (72). It has been suggested that this level of completeness is applicable for
Recently, the Writ pathway was implicated in the left-right asymmetric analyzing personal genomes in a clinical setting (72). In particular,
development of vertebrate embryos, with a role in the regulation of GC-rich first exons of genes tend to be underrepresented (33). More
ciliated organ formation and function (53-57). The key effector of complete clinical use of WGS will require higher sequencing depth, multi-
Writ signaling is P-catenin, which functions either to promote cell ad- platform sequencing and/or alignment methodologies, complementation
hesion by linking cadherin to the actin cytoskeleton via a-catenin or by exome sequencing, or all three (73). Combined alignments with two
to bind transcriptional coactivators in the nucleus to activate the ex- methods of sequencing identified —9% more nucleotide variants than
pression of specific genes (58-60). The protein that controls the switch one alone. However, these additions raise the cost of WGS, increase the
between these two processes is encoded by BCL9L (also known as time to clinical interpretation, and shift the cost-benefit balance.
BCL9-2) and serves as a docking protein to link P-catenin with other For genetic disease diagnosis, the genomic regions that harbor
transcription coactivators. BCL9L and a-catenin share competitive known or likely rlits-Aci. mutations—the Mendelianome (2, 33)-must
overlapping binding sites on (3-catenin; phosphorylation of P-catenin be genotyped accurately. In addition to exons and exon-intron bound-
determines which pathway is activated. The mutation aries, the Mendelianome indudes some regions in the vicinity of genes
found in our patients lies within the BCL9L nuclear localization signal, that have structural variations or rearrangements. NGS of genome re- n
which is essential for p•catenin to perform transcriptional regulatory gions that contain pseudogenes, paralogs (genes related by genomic
functions in the nucleus (61). duplication), or repetitive motifs can be problematic CMH064 had c c
BCL9L is one of two human homologs of Drosophila legless (Igs), a fulminant EB. Most EB-associated genes encode large cytoskektal pro-
segment polarity gene required for Wnt signaling during develop- teins with regions of constrained amino acid usage, which equate with Ic,;
ment lgs-deficient flies die as pharate adults with Wnt-related defects, low nucleotide complexity. In addition, several EB-associated genes 2
induding absent legs, and antennae and occasional wing defects (62). have closely related paralogs or pseudogenes. These features impede
Fly embryos lacking the maternal Igs contribution display a lethal seg- unambiguous alignment of short reads, which can complicate attribu- LL
ment polarity defect. BCL9L-deficient zebrafish exhibit patterning de- tion of variants by NGS. This limitation can prevent definitive exdu-
fects of the ventrolateral mesoderm, including severe defects of trunk and sion of candidate genes. For example, 45% of nucleotides in KRT14, rn
tail development (60). Furthermore, inhibition of zebrafish $-catenin an EB-associated gene, had <16-fold high-quality coverage and, thus, tel
results in defective organ laterality (54). Overexpression of constitutively may have failed to disclose a heterozygous variant. In CMH064, how- al
active P-catenin in medaka fish causes cardiac laterality defects (63). ever, this possibility was exduded by targeted sequencing of the re-
p•Catenin-deficient mice have defective development of heart, intes- gions of KRT14 known to contain mutations that cause EB.
tine, liver, pancreas, and stomach, including inverted cell types in the Furthermore, WGS is not yet effective for clinical-grade detection .2 O
esophagus and posteriorization of the gut (64). Down-regulation of of all mutation types. Copy number variations and large deletions 01
Wnt signaling in mouse and zebrafish causes randomized organ lat- require clinical validation of research methods (33). Long, simple
erality and randomized side-specific gene expression. These likely re- sequence-repeat expansions and complex rearrangements are prob- E
flect aberrant Wnt activity on midline formation and function of Kupffer's lematic. Nevertheless, with WA-type adherence to standard operation- 2
vesicle, a ciliated organ of asymmetry in the zebrafish embryo that ini- al processes, the component of the Mendelianome for which WGS is
tiates left-right development of the brain, heart, and gut (56, 65). The effective is extremely reproducible (33). Thus, the specific disns.c,
second human homolog of 1gs, BCL9, has been implicated in complex genes, exons, and mutation classes that are qualified for analysis, c?,
congenital heart disease in humans, of the type found in our patients interpretation, and clinical reporting with WGS can be precisely pre- t
(66-68). BCL9 was originally identified in precursor B cell acute lym- dieted. This is of critical importance for reporting of differential
phoblastic leukemia with a t(1:14)(q21;q32) translocation (69), linking diagnoses in the genetic disease arena Thus, although insufficient
the Wnt pathway and certain B cell leukemias or lymphomas (62). alone, rapid WGS may still be a cost-effective initial screening tool
Finally, it was recently demonstrated that the Wnt/P-catenin signaling for differential diagnosis of EB. In our study, all EB-associated genes
pathway regulates the ciliogenic transcription factor foxjla expression had >95% nudeotides with high-quality coverage sufficient to exclude
in zebrafish (57). Decreased Wnt signal leads to disruption of left-right heterozygous and homozygous nudeotide variants (>16-fold); 19 of
patterning, shorter/fewer cilia, loss of ciliary motility, and decreased these genes had >99% nucleotides with this coverage. Hence, for rig-
foxjla expression. Foxjla is a member of the forkhead gene family orous testing of all EB-associated genes and mutation types, additional
and regulates transcriptional control of production of motile cilia (70). studies remain necessary, such as immunohistochemistry, targeted se-
On the basis of this collected evidence, the symbol 1-11X6 has been quencing of unca0able nucleotides, and cytogenetic studies. Of 531
reserved for BCL9L-associated autosomal recessive visceral heterotaxy. disease genes examined, 52 had pseudogenes, paralogs, repetitive mo-
Additional studies are in progress to show causality definitively. These tifs, or mutation types that may complicate WGS for comprehensive
findings support clinical WGS as being valuable for research in reverse- mutation detection. The comprehensiveness of WGS will be enhanced
translation studies (bedside to bench) that reveal new genetically ame- by longer reads, improved alignment methods, and validated algo-
nable disease models. rithms for detecting large or complex variants (2, 4).
Finally, in singleton (sporadic) cases, such as CMH064, family his-
Addressing limitations tory is often unrevealing in distinguishing the pattern of inheritance.
In one remaining prospective patient, rapid WGS failed to yield a potential For example, inheritance of ES can be dominant or recessive. Of two
or definitive molecular diagnosis. Currently, WGS cannot survey every plausible heterozygous VUS detected in candidate genes in CMH064,
wwwScienceTranslationalMedicine.org 3 October 2012 Vol 4 Issue 54 154ral 3S 7
EFTA00315107
RESEARCH ARTICLE
one was a de novo mutation in connexin 26, which is associated with to individual clinical presentations, in accord with published guide-
KIDS, that can be fatal in neonates (43,44). The phenotype of C.MH064 lines for genetic testing in children and with NGS (9, 25, 28). SSAGA
was not typical for IUDS, and functional studies are in progress. For eval- has a menu of 227 clinical terms, which are arranged in nine categories
uation of dominantly inherited diseases, WGS requires that the parents (fig. SI). SNOMED CT (Systematized Nomenclature of Medicine—
be concomitantly tested either by rapid WGS, by exome sequencing, or Clinical Terms) (29) map to 591 well-established recessive diseases with
by resequencing of candidate de novo variants. known causal genes (table SI). Phenotype-to•disease-to•gene mapping
Rapid WGS failed to yield a definitive molecular diagnosis for was informed by Gene Reviews (30), OMIM Clinical Synopsis (10),
CMH076. No known mutations were found in 89 disease-associated MitoCarta (45), and expert physician reviewers.
nuclear genes or the mitochondrial genome. 'This was an important Upon entry of the features of an individual patient, SSAGA nomi-
negative finding because a molecular diagnosis of several of these nates the corresponding superset of relevant diseases and genes, rank
genes is "actionable." That is, specific treatments are indicated (such ordered by number of matching terms (fig. SI). It also contains a free-
as pyruvate carboxylase deficiency, thiamine responsive congenital ac- form text box that allows physicians to enter findings for which no
idosis, biotinidase deficiency, fructose 1,6-bisphosphatase deficiency, SNOMED term exists, clinical term qualifiers, relevant family history,
and coenzyme Q10 deficiency). Likewise, exclusion of actionable and specific genes of interest. The diagnostic sensitivity of SSAGA im-
diagnoses can prevent empiric institution of inappropriate treatments. proves with use, by manual updating of mappings in cases where nom-
Exdusion of known genetic diseases from a differential diagnosis is inations failed to include the causal gene. SSAGA is extensible to
also of psychosocial benefit to family members and assists in guiding additional diseases, genes, and clinical terms. Interpretation of results n
physicians regarding additional testing. There were no VUS with suitable was manual on the basis of ranking of variant reports yielded by RUNES
inheritance patterns, in CMH076 or in either of the healthy parents, in on SSAGA-prioritized candidate genes, supplemented with expert gene c ;
known disease genes or in the remaining 929 nuclear-encoded mitochon- nominations (fig. S2). In some pedigrees, the presumed pattern(s) of
drial genes (45).
In contrast to the rapidly declining cost of WGS, the computation-
inheritance allowed additional variant ranking based on obligatory
genotypes in affected and unaffected individuals. Aligned sequences 2
m
al cost of genome analysis is largely governed by Moores law (74). containing variants of interest were inspected for veracity in pedigrees
Sequence alignment, variant calling, and genotyping took 16 hours. with the Integrative Genomics Viewer (32). LL
Extremely rapid WGS is of practical use in clinical guidance only
when married to equally rapid, cost-effective, deployable, and facile Genome and exome sequencing rn
interpretation and analysis (2, 4). We are continuing to improve the Isolated genomic DNA was prepared for rapid WGS with a modifica-
speed of sequence base calling, alignment, and variant calling. It is tion of the Illumina 'auSeq sample preparation (Illumina). Briefly, 500 ng
likely that this interval can be halved and that HiSeq 2500-based rapid of DNA was sheared with a Covaris S2 Biodisruptor, end-repaired,
WGS can be performed in fewer than 36 hours by the end of 2012. A-tailed, and adaptor-ligated. Polymerase chain reaction (PCR) was C u
Clinical validation of rapid WGS, however, will take some time. omitted Libraries were purified with SPRI beads (Beckman Coulter). T.) a)
Quantitation was carried out by real-time PCR. Libraries were dena-
tured with 0.1 M NaOH and diluted to 18 pM in hybridization buffer. E
MATERIALS AND METHODS Samples for rapid WGS were each loaded onto two flowcells,
E
followed by sequencing on alumina HiSeq 2500 instruments that were
Consent set to rapid run mode. Cluster generation, followed by 2 x 100 cycle
This study was approved by the Institutional Review Board of CMH. sequencing reads, separated by paired-end turnaround, were performed cu
Informed written consent was obtained from adult subjects and automatically on the instrument. cu
parents of living children. Isolated genomic DNA was also prepared for Illumina TruSeq exome
or custom gene panel sequencing with standard Illumina TruSeq pro-
Case selection tocols. Enrichment for the custom gene panel was performed twice by 8
CMH is a nonprofit children's hospital with 314 beds, including 64 level 4 Illumina hybrid selection with 20A77 eighty-nucleotide probes for 8366
NICU beds. It provides 48% of neonatal intensive care in the Kansas City genomic regions, representing exons and 20 intron-exon boundary
metropolitan region. In 2011, the NICU had 86% bed occupancy. Retro- nucleotides. It encompassed 2,158,661 bp, 525 genes, and 591 reces-
specthe samples, UDT002 and UDT173, were selected from a validation sive diseases (2, 33) (table SI). The probes were designed to target 350
set of 384 samples with known molecular diagnoses for one or more nucleotide genomic targets, with an average density of 14 probes per
genetic diseases. Seven prospective samples were selected from families target (range, 2 to 56). Custom gene panel-enriched samples were se-
with probands that presented in infancy, among 143 individuals with- quenced on HiSeq 2000 instruments with TruSeq v3 reagents to a
out molecular diagnoses who were enrolled between 22 November 2011 depth of >3 GB of singleton 100-bp reads in samples UDT173 and
and 4 April 2012 for exome or genome sequencing. UDT002, respectively; 319 and 38.3% of base pairs were on target
defined with a 0-bp extension, representing 469- and 501-fold enrich-
Clinicopathological correlation and interpretation ment in samples UDT173 and UDT002, respectively. Exome-enriched
The features of the patients' diseases were mapped to likely candidate samples were enriched twice with standard Illumina hybrid selection
genes. In part, this was performed manually by a board-certified pe- and were sequenced on HiSeq 2000 instruments with TruSeq v3 re-
diatrician and medical geneticist. In part, it was performed automat- agents to a depth of >8 GB of singleton 100-bp reads per sample.
ically by entry of terms describing the patients' presentations into a Genome and exome sequencing were performed as research, not in
new clinicopathological correlation tool, SSAGA (2). It was designed a manner that complies with routine diagnostic tests as defined by the
to enable physicians to delimit WGS analyses to genes of causal relevance CLIA guidelines.
wwwSdenceTrandationalMedicine.org 3 October 2012 Vol 4 Issue 54 154ral 3S 8
EFTA00315108
RESEARCH ARTICLE
Sequence analysis Patient 1
CASAVA 1.8.2 (alumina) performed gapped ELAND alignment of CMH064 was a male bon at 33 weeks gestation with erosive dermatoses.
HiSeq 2500 sequences to the reference nuclear and mitochondrial ge- He was delivered vaginally following induction for preedampsia. Des-
nome sequences llig19 and GRCH37 (NC_012920.1), respectively] as quamation and erythrodemia from the hairline to occiput were present
wen as variant identification. HiSeq 2000 sequences were aligned to at birth (Fig. 2A). Denuded, hyperpigmented, and partially scarred le-
the reference nuclear and mitochondrial genome sequences with GSNAP, sions were noted above the upper lip, over the mentum, and in place of
and variants were identified and genotyped with the GATK (234-36). eyebrows. He had a truncated foreskin. His nails were dystrophic and
Sequence analysis used base-call files, FAS'I'Q files that contain sequences yellowed. There were no vesicles, pustules, blisters, or mucosal lesions.
and base-call quality scores, the compressed binary version of the Se- Family history was positive for psoriasis. His mother had a healthy
quence Alignment/Map format (a representation of nucleotide sequence daughter from a prior union; there was no history of fetal loss. His
alignments), and Variant Call Format (a format for nucleotide variants). father was healthy.
Nucleotide variants were annotated with RUNES (2), our variant char- Cultures and herpesvirus PCR were negative. He developed severe
acterization pipeline, which incorporated VEP (Variant Effect Predictor) neutropenia by day 3. Skin sloughing worsened Rigid bronchoscopy
(37), comparisons to NCBI dbSNP, known disease mutations from the and intubation were necessary because of fibrinous oropharyngeal
HGMD (38), and additional in silico prediction of variant consequences exudate.
with ENSEMEL and UCSC gene annotations (39, 40) (fig. S2). RUNES Skin biopsy histology revealed acantholysis, loss of cohesion be-
assigned each variant an ACMG pathogenicity category (25-27) and tween keratinocytes, and empty lacuna (Fig. 3A). There was focal e
an allele frequency on the basis of 722 patients sequenced since October dermal infiltration with neutrophils and lymphocytes and complete
2011. Rapid WGS in CMH064 and exome sequences of his parents sloughing of the epidermal layer with focal defting at the suprabasal
were also analyzed by Clinical Sequence Miner (deCODE Genetics). layer (Fig. 3B). Immunolluorescence staining was negative for IgA
it
in
2
.o
w
u_
a
O
2'
O
cn
id
E
8C
0)
E
icr)
E
co
co
3
Fig. 2. Skin lesions in patient CMH064. (A) Desquamated lesions with
erythroderma on the scalp at birth. (B) Day 30 progression of desquamatIon.
His fingers were edematous and discolored and had retained only three nails.
www.ScienceTranslationalMedicine.org 3 October 2012 Vol 4 Issue 1S4 154ra13S 9
EFTA00315109
RESEARCH ARTICLE-
'Willi s ,
4 ... ir l g. - ..
i, %A %, •
- -.
iel itt./.47,t
•
1,;•,1A
lisi e.. ter.,6 c
4,44 a. Af?.... arcs._ • ...
4*, f , •
„.
0 la
.. ..„, .. t. 17404
% • r .....•-:.:-.% 4‘
IN ). • •
, 1 .. -• •.. .•r a. 1,,,. $ .
,. ,c,... r ,7„. , I •
•,, ‘
7:I:
• * 4 t.s4:sil;
11.:`:::11i...
. . ... , • •
• .
A
,
19 I i 4 • 1 ' i 4 c .
„ - le A 4
a . n
r . . . ,
e tr,..e.ai• ' 4 or. .-
.., .. 1;14;SP/ro
se! j....4,,,?es . ••••
., •
.. 7 II ' li ' le , • .:, •f.. '
'N
ii .
:., , . 0
• t.•• t.7
•• 'r 16,• 4;•; 4a.;•4 .- •
a ••
.* • . O
, .41.0.• ..,1•91,.... •
Pilo %teas. • ' Y. «Ili 2
Fig. 3. Skin lesion histology for patient CMH364. (A) Dermal acantholysis mal infiltration of neutrophils and lymphocytes (double.headed arrow). The
(loss of intercellular connections resulting in loss of cohesion between ke- epidermal layer shows complete sloughing with focal clefting at the LL
ratinocytes) and formation of empty lacunae (cavities; arrow). (In Focal der- suprabasal layer (arrow). C
a
Fr
(immunoglobulin A), IgM, and IgG except for linear staining for whose pregnancy was notable for decreased movements at 35 wedcs eel
C3. Additional skin immunofluorescence studies revealed slightly gestation. His mother and father were healthy. Variable decelerations E
reduced plakoglobin and desmoplakin and normal laminin 332; col- in heart rate were noted on the day before delivery. labor was compli- C S
lagen types 4, 7, and 17; and plakophilin-1. Electron microscopy con- cated by prolonged rupture of membranes, and delivery was by vacuum
firmed the absence of dermoepidemtal junction (DEJ) separation and extraction for meconium staining. Apgar scores were 2, 3, and 5 at 1, 5,
showed focally widened spaces between keratinocytes and cell vacuol- and 10 min, respectively. He had poor respiratory effort and hypotonia E
ization from the DEJ to the stratum corneum. Hemidesmosomes were and required intubation. Upon transfer to CMH on day 2, he had lactic
E
normal. Some keratinocytes had large solitary vacuoles, abnormal con- acidosis (lactate, 12 mmolldl), coagulopathy, and cloudy corneas.
densation of keratin filaments, and perinuclear pallor. Some desmosomes Multiple cultures were negative. Echocardiogram showed chamber en-
had ragged edges. There were no intracellular indusions. Negative labo- largement, reduction in biventricular ftinctkm. noncompaction cardiomy-
ratory studies included karyotype, Ro, La, Smith, ribonudeoprotein, and opathy, mild tricuspid insufficiency, and mild aortic insufficiency. Urine
Sd-70 autoantibodies. Igs were unremarkable apart from an elevated testing revealed normal amino acids and elevated 3-methyglutaconic z
serum IgA. acid, 3-methylglutaric acid, and 2-ethyl-3-hydroxy-propionic acid Long-
Sloughing of the skin, mucosal surfaces, and cornea continued to chain fatty acids, acyl-camitines, lysosomal hydmlases, li-galactosidase,
worsen, and by day 30, his activity level had markedly decreased (Fig. 13-glucumnidase, sphingornyelinase, glucocerebrosidase, a- t•iduronidase,
2B). His fingers were edematous and discolored and had retained only and a -glucosaminidase were normal Pressors were required for hypo-
three nails. On day 39, he developed purulent drainage from facial tension, and acidosis increased He was diagnosed with hypoxic ischemic
lesions. Skin cultures were positive for Escheridtia coil and En terocoaus encephalopathy. On day 3, lactate was 282 inmol/d1. On day 5, respi-
faecalis, and blood cultures for E coll. Antibiotics were administered. He ratory distress worsened, accompanied by bloody endotracheal secretions;
was thrombocytopenic and anemic, necessitating numerous transfu- arterial pH was 7.04 and lactate was 22.0 mmolldl. Medical interven-
sions. On day 47, ultrasound revealed nonocdusive portal vein and left tions were withdrawn at the family's request, and he expired on day 5.
brachiocephalic vein thrombi. By day 54, he developed metabolic ad- Postmortem testing by array comparative genomic hybridization (aCGH)
dosis, bloody stools, and persistent tachycardia. Medical interventions and sequencing for mitochondrial transfer RNAs and TAG, associated
were withdrawn and he died on day 54. At autopsy, suprabasal acan- with Barth syndrome, were normal.
tholysis was present in the skin and the esophageal mucosa. Dideoxy
sequencing of candidate genes KRIS, ESP, JUP, TP63, and KRTI4 exons Patient 3
I, 4, and 6 (the regions harboring most KRTI4 mutations) was negative. CMH172 was a female with intractable epilepsy. She was delivered at
39 weeks gestation by Cesarean section after an uncomplicated preg-
Patient 2 nancy. No exposure in utero to drugs, alcohol, or medications was re-
CMH076 was a male born at term with lactic acidosis, cardio- ported. Birth weight was normal, length was 46 cm (.3%), and head
myopathy, and corneal clouding. He was born to a primigravid mother circumference was 33 an (<3%). Amniotic fluid was meconium-stained
www.SeenceTranslauonalMedicine.org 3 October 2012 Vol 4 Issue 154 154ra135 10
EFTA00315110
RESEARCH ARTICLE
Apgar scores were 6, 7, and 8 at 1, 5, and 10 min, respectively. Family Table Si Disease genes nominated by SSAGA on the basis of the clinical features of patient
UDT002.
history was positive for a female cousin with profound intellectual disabil-
Table 54. Disease genes nominated by SSAGA on the basis of the clinical features of patient
ity and infrequent seizures, and two cousins by a consanguineous marriage UDT173.
who died at 6 and 8 weeks of age of intractable epilepsy; all were from Table 55 Disease genes nominated by SSAGA on the basis of the clinical features of patient
the same small Mexican town as the proband. Seizures started 1 hour 041064.
after delivery. Antibiotics were given empirically until cultures and cere- Table 56 Candidate genes for EB with pseudogates. paralogs, or segments with reduced
sequence complexity.
brospinal fluid (CSF) herpesvirus PCR returned negative. Seizures Table 57. Disease genes nominated by SSAGA on the basis of the clinical features of patient
continued despite multiple antiepileptic medications. CSF (including CA*1076.
glycine level and CSF/plasma ratio) and brain MR1 were normal. Table SR Disease genes nominated by SSAGA on the basis of the clinical features of patient
Electroencephalogram (EEG) showed focal epileptiform and sharp 041172.
Table S9. Disease genes nominated by SSAGA on the basis of the clinical features of patient
wave activity. Blood ammonia, electrolytes, pH, and glucose were 041184.
normal. Oral feeding was poor. She was intubated and required
increasing respiratory support for low Sa02 and bradycardia. Oph-
thalmologic examination and radiologic skeletal survey were normal.
An echocardiogram revealed a patent foramen ovale, tricuspid regur- REFERENCES AND NOTES
gitation, and peripheral pulmonary stenosis. Her karyotype was 1. E. D. Green, M. S. Guyer: National Human Genome Research Institute, Charting a course for
normal. aCGH was not diagnostic, but multiple tracts of homozygosity generic medicine from base pairs to bedside. Abjure 470, 204-213 (2011).
suggested shared parental ancestry. A repeat brain MR1 at age 3 weeks 2 & F. IGinguncre, D. L Dinwiddie. N. A. Miller. S. E. Soden, C J. Saunders. Adopting orphans: ai
was normal. Upon transfer to CINH at 5 weeks of age, she was small Comprehensive genetic testing of Mendelian diseases of childhood by nextgeneration
but symmetric, with bitemporal narrowing, micrognathia, flat nasal sequencing. Expert Rev. Md. Oscan. 11, 855-868 (2011).
I M. N Banbridge. W. Womiewsk D. R. Murdock l. Rierimm,C Gonzsgeleuregui, I. Newham. Z.
bridge, upslanted palpebral fissures, uplifted ear lobes, redundant J. G. Reit A K Rnk, M. B. Varga; M.0 Cingras, CIPA 'Aunty, L D. Heang. 5. YousaLL R. Lupdri. 01
helices, and fifth finger clinodactyly. She had hypertonia, persistence P. A Cibbs,Wlidegenome sequencing for optimized patient L ScL Trans( Med. 3, 2
of cortical thumbs, hyperreflexia, donus, and facial twitching. B6 chal- 87re3 (20113
lenge improved her EEG transiently, followed by return of multifocal 0. S. F. Kingsmore. C. J.Saunders, Deep sequencing of patient genomes For disease diagnosis: LL
When will it become routine? Sri. Trend. Med 3, 87P523 12011).
sharp waves. Serum amino acids and urine organic acids were normal. 0
S. C G:nzaga-lauregui, 1. R. Lupski, R. A. Gibbs, Human genome sequencing in health and
Recurrent seizures continued both clinically and by EEG. After lengthy disease. Anna. Rev. Afed. 63. 35-61 120121.
0
discussion, the parents requested withdrawal of support. 6. A R LupskL 1. W. Beknont. E BoervaMde, R A. Gibbs, Clan genomics and the complex ci )
architecture of human disease. Celt 147. 32-03 (20111. 01
Patient 4 7. J. R. Lupski, J. G. Reid. C. Gonzaga-laureguit, D. Rio Deiros, D. C. Y. Chen, L. Nazareth, E
CMHI84 was a male with visceral heterotaxy and congenital heart M. Bainbridge, H. Dinh, C. ling, D. A. Wheeler, A. L. McGuire, F. Zhang, P. Stankiewia, 8
disease (dextro-transposition of the great arteries, total anomalous pul-
J.1. HaJperin, C Yang C. Gellman, D. Guo, R. K. Irikat, W. Tom, N. J. Ninths, D. M. Many.
P. A. Gibbs, Wbolegenome sequencing in a patent with Charcot-Marie-Toah Neutpathy.
N
monary venous return with pulmonary veins connecting to the right Engl. 1 Med. 36Z 11B1-1191 (2010).
atrium, a large ventricular septal defect, pulmonary valve and main & E A. Wortley, A. N. Meyer. G. D. Syvenon, D. Hetling & & Bonacci, B. Decker, J. At Serpa.
T. Clinu,PA R. Tschannem R L Veith, M.1. Basehore, U. Broedo& A Tomita-Michel. M.l. Arca. M1
pulmonary artery atresia, mildly hypoplastic branch pulmonary ar- J. T. Casper, D. A. Margin, D. P. Rick, M.I. Hewer, 3. M. Routes, 1. W. Verbsky. H. 1. Jacob. E
teries, patent ductus arteriosus with ductal-dependant left-to-right O. P. Climmcck, Making a definitive diagnosis: Successful clinical application of whole 2
flow, and large atrial septal defect with obligate right-to-left flow). exam sequencing in a child with intractable inflammatory bowel disease. Genet Med .2
There was sites inversus of the spleen, liver, and stomach, with the 13. 255-262 (2011).
aorta on the right of the spine and inferior vena cava on the left. Family 9. American College of Medical Genetics and Ceramics (ACMG). Policy Statement. Points to
Consider in the Clinical Application of Genetic Sequencing, IS May 2012; available at a
history was positive for a 6-year-old brother (CMH185) with the same
findings (dextrocardia, ventricular inversion, double outlet right ventricle,
http://wwmacmgnet/StaticContentIPPGrOinial_Applcaten-M-Denornic-Secluenoirlicil.
10. Online Menddian Inheritance in Man. fricKusickkathans Institute of Genetic Medicine. 0
5
pulmonary stenosis, small pulmonary arteries, interrupted inferior vena Johns Hopkins University (Baltimore, MD): evadable at www.mnimorgrstatistics (accessed CI
cava with amous continuation, and sites inversus of the liver and spleen). 13 July 20121.
II. F. R. Hauck, K. 0. Tanabe, R. V. Moon, Racial and ethnic disparities in infant mortality.
His parents (mother, CA111186, and father, CMH2O2) and two other sib-
Sent Perinatal 35, 209-220 (2011).
lings (one male and one female) were healthy. Testing of ZIC3, associated II M. C. Lynberg, M. J. Khoury, Contribution of birth defects to infant mortality among
with X-linked recessive heterotaxy I, was normal. Patient 4 remains in the racial/ethnic minority groups, United States. 1983. MMWR CDC Surveil. Samm. 39, 1-12
NICU and is undergoing cardiac surgery. (1990).
II K. D. Kochanek, S. E. ICameyer. 1. A. PAartin, D. M. Strobino, & Guyer, Annual summary of
vital statistics: 2009. Pedoteks 129, 33/3-308 120121.
10. D. Alexander, 1. W. Hansom WHO research initiative in newborn screening Ment Retard.
SUPPLEMENTARY MATERIALS Dev. Mani Res. Rev. 11, 301-304 120061.
IS. American College of Medical Genetics' Newborn Screening Expert Group, Newborn
vniv.sciencetranstatanalmeckine.org/cgi/contentAull/4/150/154m135/DC1 screening: Toward a unilomv screening panel and system. Executive summary. Genet
Materials and Methods Med. It, 1S-11S (2006).
Fig. SI. Candidate gene selection by SSAGA for automated variant characterization and inter- I& M. L Come, A. Bana, M.D. Bayed; A. Perez.Muriuzuri, 1. R. Fernandez-Lorenzo, J. M. Frage.
pretation guidance. Inborn errors el metabolism in a neonatology unit bnpact and longterm results. Palter. krt.
r9 Si Automated variant thwack/intim by RIJNES. 33, 13-17 (2011).
Table SI. Five hundred ninetyone recessive cisesses and genes for which clinical terms and a I?. G. .I. Downing, & E. Zuckerman, C. Coon M. A. Uoyd.Puryear. Enhancing the quality and
targeted enrichment panel KMH-Clx1) hwe been developed. efficiency of newborn screening programs through the use of health information technol.
Table S2. Comparison of the rapid WG5 data set from sample 0614064 with three different ogy. Sernin PerMotal 34, 156-162 (2010).
afignment and variant detection methods (GSNAP/GATIL the Illumine CASAVA alignment tool, It J. D. Lantos, W. L Meadow, Costs and end-atilt are in the NICU: Lessons for the PHU?
and BWA tool). J. Low Med. Wks 39. 194-200 (20111.
www.ScienceTranslationalMedicine.org 3 October 2012 Vol 4 Issue 154 154o135 11
EFTA00315111
RESEARCH ARTICLE
19. B. 'Merrell F. Lorey, R. Fawn. 0. Frazier, G. Hoffman C Boyle. a Green, 0. Devine, H. Hannay H. Orman G. P. Barber, D. Ilausskr, W. lames Kent, The uac Genome Browser database:
Impact of Expanded Newborn Screening—United States. 2006 MMWR Mot Meet Wkly. Extereions and updates 2011. Hickok Acids Res 40, D918-0923 (2014
Rep. 57, 1012-1015 12004 41. M.l. Clark. R. Chen. It Y. Lan, K J. Karczewski, R. Chat G. Euskircluen, A 1. Butte, M. Snyder.
20. Health Resources and Services Administration, Secretary% Advisory Committee on Herita- Performance comparison of exeme DNA sequencing technologies. Not Riorechnot 29.
ble Disorders in Newborns and Children. 2011 Annual Report to Congress, Rockville MD 908-914 (2011).
(2011); htleliwwwhnagov. 42 R. G. Cotton, C R Striver. Proof of 'disease causing' mutation. Nam. Mums. 12.1-3
21. G. Pfeffer, K. Majamaa, 0. M. Turnbull. D. Therburn, P. F. Chinnety, Treatment for mitechon- (1998).
dial disorders. Cochrane Database Syst Rev. 4, CD004426 (2012). 41 G. Richard Connexin dsaden of the skin. Can. Demoted. 23, 23-32 120051.
22. R. H. Singh, F. Rohr, P. L Splett Bridging evidence and consensus methodology for in- 44. E Sbisfran. D. Fddmann, I Bengoa, S. Freitag. V. Abade, Y. de PIOU.0 Oedema. S. Hadj-Ratia.
herited metabolic dsorden: Creatng nutriticn guidelines A £vd CM. Prat 10.1111/j.1365- Germane mesakism in keratiris-ichthyosiscleathess syndrome:Pre-natal diagnosis in a familial
2751201101807x (2011). lethal faro. an. Genet. 77, S87-592 00101.
23. N. L Sebreira, E. T. Cirulli, 0. Avramopoulos. E Wohler, G. L. Oswald, E. L Stevens, D Ge, 45. D. J.Pagkarini. S. E. Cahn B. Chang. S.A. Shah, S. B. Vafai„ S. E. Ong G. A. WaVerd, C.Sucjana.
K. V. Shianna, J. P. Smith. J. M. Maia, C. E. Gumbs, J. Penner, G. Thomas, D. Valle, A. Bench. W. K Olen. a E Hill M. Vidal 1.G. Evans, D. R. lhorbum. S. A Carr, V. Kabala, A
J. E. Hoover-Fang. 0.8. Goldstein, Whole-genet-Ise sequencing of a single Frahled together mitechoridrial protein compendium ducilates complex I disease biology. Gad 134 112-123
with linkage analysis identifies a Mendelian disease gene. PtoS Genet 6 e1000991 0010).
24. K A. Wetterstrand. DNA Sequencing Costs: Data from the NHGRI Large-Scale Geneme 46. E G. Puffenberga. R.N. /inks, C Soignee, K Cibultkis R. A. Willed N. P. Achily, R P. Cassidy.
Sequencing Program; available at www.genomegovisequentingcosts (accessed 13 July C 1. Forentini„ K F. Heiten.1.1. Lawrence, M. H. Mahoney, C I Willer, D. T. Nair. K. A. Politi.
20121. K N. Worteskr,RA Setton, R Didazra E A Sherrnartl T. Eastman C Francklyn S. Robey6or-4
25. C. S. Richards, S. Bale, D. B. Bellissimo, S. Das, W. W. Grady, M. R. Hegde, E. Lyon, N. L Rider. S. Gabriel, D. H. Morton, K A Strauss, Genetic mapping and exome sequencing
B. E. Ward: Molecular Subcommittee of the ACMG Laboratory Qinlity Assurance Committee, identify varianh associated with five novel diseases. PtoS One 7, e28936 (2014
ACMG recommendations for standards for interpretation and reporting of sequence 47. E Teed, The [reader Destruction a/Meth:int (Basic Books, Perseus Books Group, New York, :LP
variations: Revisions 2007. Gant bled 14 294-3C0 (2C08).
2Q A. Maddalena S Bale. S. Oat, W. Grady. S. Richards: ACMG Laboratory Quality Assurance
NY, 2012).
48. S. E. Barauini„ J. Mudge, l C van Velkinburgly P. Ithankhanian, I. Khrebtukova, N. A. Miller,
N
Committee, Technical standards and guidelines: Molecular genetic testing for ultra-rare L Dung& D. Farmer, C J. Bell, R.W.Kim.G. D. May, l E.Woodward. S./Caller. J. P.MtElroy,
disorders. Genet. Med. 7, 571-583 (2005). R. Gomez. M. J. Paide, L E Oendenen. E. E. Ganusova, F. 0. Sdakey, T. Ranwraj. 0. A. Man,
27. M. A. Zoccoli. M. Chan, J. C. Erker, A Ferreira-Gonzalez, I. M. Lubin, Nucleic Acid Sequenc- J. J. Huntley. S. Luo. P. Y. Kwok, T. D. Wu, G. P. Schroth, J. R Oksenberg, S. L Hauser. m
ing Methods in Diagnostic Laboratory friedeine Approved Guideline. NCCLS document S. F. Kingsmore G2•10111.• epigenome and RNA sequences of morozygetic tarn discordant 2
MM9.& NCCLS. PA, USA (2004). For mdtkle sclerosis. hbture 464, 1351-1356 (20103
28. American Society of Human Genetics Board of Directors; American College of Medical
N
49. 1. S. Kahane, 0. R. Mast's, R & Altman. The incidentaleme A threat to genernit medicine. U.
Genetics Board of Directors, Points to corakler. Ethical, legal, and psychosocial implica- JAM 296.212-215 (2006).
tions of genetic testing in children and adolescents, Am. A Hum Genet. S7, 1233-1241 C A Casa, S. K Savage. P. L. Taylor, R. C. Green. A. L. McGuire. K. D. Mandl. Disclosing °
0995). pathogenic genetic variants to research participants: Quantifying an emerging ethical re- ,..2)
29. httpc/AnvernIrruiihgovhesearch/umb/Snaned/snerned_mainhtml. sporistity. Genorne Res 22,421-428 (2012).
30. GerieReriews at GeneTests: Medical Genetics Information Resource (database online). Uni. St P. V. Asharani, K. Keupp, O. Semler, W. Wang, Y. Li, H. Thick, G. Vigil, E. Pohl, 1. Becker. 0 m
varsity of Washington, Seattle (1997-2011): available at httpor/wwwgenetests.org. P. Frommdt,C Senates J.Mtmaller, K Zintrnermenn D.S. Greenspan. N. A. Akarsu, C. Netzer. E
31. B. Ewing. P. Green, Base-calling of automated sequencer traces using phred. II. Error prob- E Schen.% R. Wirth, Al. Hammenehrniit P. Numbers B. Weiner, T. 1. Camey, Attenuated 8
abilities. Gnnome Res. 8,186-194 11998). BMPI function compromises osteogenesis. leading to bone fragity in humara and setnfish.
32. J. T. Retire's:n.14. lbervaldsdetti, W. Vrindler, M Guttman E S Lander, G. Getz, J. P. Maim.. Am. I Hum. Genet. 90.661-674 0018
N
Integrative geromics viewer. Not. BiOeCchttli 29,24-26 12011). 52 1. H. van Es. N. Barker, H. Clever', You Wnt some, you lose some: Oncogenes in the Wnt
33. C J. Bell, D. L. Cinwkfdie, N. A. Miller, S. L Hateley. E. E Ganusova. J. Mudge, R. J. Langley, signaling pathway. Carr. Opin. Genet. Dev. 13,28-33 12003).
L Dans C C. Lee. F. D. Schilkey, V. Shah, 1. E. Woodward. H. E. Pelham, G. P. Schroth, SI K. M. Cadigan, R. Nusse. Wm signaling: A common theme in animal development Genes CO
R. W. IGin, S. F. Kingsmore, Carrier testing for severe childhood recessive diseases by Dev. 11. 3286-3305 0998
next-generation sequencing. Sri. Trans. Med. 3, 65ra64 (2011). 54. M. Bong J. Zhas S C Lin. A. Marg. II-Catenin I and IScaterin 2 play similar and distract roles 2
34. T. O. Wµ S. Nacu, Fast and SNP-tekrant detection of complex variants and splidng in short in left-right asymmetric development of zebrathh embryos Development 139,2009-2019 7;
reads. Edinformoria 26,873-88I (2010). (2012). ffi
35. A. McKenna, M. Hanna. E. Banks, A. Sivadsenko. K Cibulskis, A Kernytsky, K. Garimella, St L Schneider, P. N. Schneider, S. W. Deify. & Lit L 1. Barton. T. Wesdall. D. C Slusarski. Tol
D. Altshukt, S. Gabriel. IA Daly. IA A DePriste, The Geneme Analysis Toad: A MapPeduce Zebrafish Nkdl promotes Dvl degradation and is required for left-right patterning. Dev. Biol. 0
framework For dialyzing next -generation DNA sequencing data. Groan Res. 20. 1297-1303 344 22-33 ROM
(20101. 56 X. Lin X Xu, Distinct functions ef Wn1/6-catenin signaling in Ks/ development and card& g
36. M. DePristo, E. Banks, R. Poplin, K. V. Garimella. 1. R. Maguire. C Hartl, A. A. Philippakis, asymmetry. 134 207-217 (2009).
G. del Mg.* M. A.Rivas. At Fawns, A Mdknna, t J. Fennell, A M.Kernybky. A. Y.Svachenko, 57. A. Caron, X. Xu, X. Un, Witt/fircatenin signaling &reedy regulates Fowl expression and
K. Gbulskis S. B. Gabriel, a Aftshukr, M. 1. Daly. A framework For variation dacovery and Oliegenesis in zebralish Kupffers vesiek. Development 139,514-524 120121.
genotyping using next-generation DNA sequencing data. Not. Grad. 43. 491-498 (2011). SR R. Keller, From cadherins to catmint: Cytoplasmic protein interactions and regulation of
37. W. McLaren. & Pritchard, D. Rios, Y. Chen P. Find, F. CunninghamDeriving the consequences cell adhesion Trends Genet. 9,317-321 (1993).
of genornic variants with the Ensembl MI and SNP Effect Predictor. Eaoinformatics 26, 59. K. Korinek. N. Barker, P. J. Morin, D. van WkIsen. R. de Weger, R. W. Kinder. B. Vegelstein.
2069-2070 12010). H. Owen Constitutive transoiptional activation by a 6-caterin.Td complex in Mir colon
38. P. D. Stenscn. E. V. Ball, K Howells, A. 0. Phips, M. Mort, a N. Cooper, The Human Gene carcinoma Science 274 1784-1787 (19971.
Mutation Database Providing a comprehensive central mutation database for molecular 661. Behrens, 1. P. von Kries, M Kahl. L Bruhn, 0. Wedlith, R Gresschedl. W. Birehmeier.
diagnostics and personalized genomies. Han. Genoa*, 4,69-72 (2003). Functional interaction of 13-takenin with the transcription factor LEF-I. Nature 382, 638-642
39. P. Flitek, M. R. Amode, a Barrett. K. Beal, S. Brent. O. Carralho-Sava, P. Oapham, G Coates, (19961.
S. Fairley, S. Fitzgerald, L. Gil. L. Gordon,M. Hendrix, T. Hourlier, N. Johnson. A K KahAri, 61. F.It Brembeck 7. Schwarz.Romend, J.Bakkers,& Wilhelm, M. Ibmmersehmidt, W. Birthmeia.
0.Keats. S. Keenan R IGnsella.M. Komrdevaka,GAescielrry,E Kulethe, P. Lansay I. Lagden, Essential role of BCL9-2 in the switch between 13-catenires adhesive and transcriptional
W. McLaren, M. Mullet*, B. Overduin, M. Pignatelli, B. Pritchard. H. S. Riat, G. R. Ritchie, functions. Genes My. 18, 22B-2230 12004).
M. Ruffier.M.Schuster, D Sabra/ Y. A. Tang K Taylor, S. Trevanien, J. Vandrevcova, S. White, 62 T. Kramps. 0. Peter. E. Brunner, D. Hellen, B. Froesch. S. Chatterjee, M. Murone, S. Zollig.
M. Wilson, S. P. Wilder, B. L. Men, E. Birney. F. Cunningham, I. Dunham. R. Durbin, K Basler, 1NntAvirigkss signaing requies BCLarlegessmoiated naraitment of pygopus to
X. M. Fernandez-Suarez, J. Harrow. J. Herrero, T. J. Hubbard, A. Parker, G. Proctor, the nudes A-GOICIIII-TCF complex. Cell 109. 47-60 (20021.
G Spuefick J. Vogel, A. Yates, A. Zadissa. S. M. Searle, Ensembl 2012. Mnfek Acids Res. 40, 61 & Bajoghk N. AghaaArei. D. Scolded, T. Czerrry, The roles of Grouches/71e in left-right
084-090 420121. asymmetry and Kupfter's vesiek erganogenesis. Dev. Bid. 303, 347-361 12007).
40. T. Ft. Dram, EtKarolthit A. S. Zweig. A S.linrichsB.1.Rawy,RIA Kuhn, LR Meyer, ht Wong. 64. T. Grigeryan P. Wend, A. Klaus, W. Efirchmeier. Deciphering the function of canonical Wnt
C.A Sloan, K R Rosenbloom, G Rex, & Mead A. Pohl, V. 5.4.4allaci,C H.L.I,K Leansed,V.Kirkup, signals in development and disease: Conditional loss- and gain-of-function mutations of
F. Hsu, Ft & ManeL Guntradco,1W6dt:et B.M. Giardne. P. A FujitaM tick's's.M. S. Clot. 3-catenin in mice. Genes Dee 22. 2303-2341 (20081.
www.SciencerranslationalMedicine.org 3 October 2012 Vol 4 Issue 154 154o135 12
EFTA00315112
RESEARCH ARTICLE
65. M. A. Nakaya. K. Biris. T. Tsukiyama, S. Jaime, J. A. Rawls, T. P. Yamagodi, Wnt3a links left• on W.
76. C S. Richards. S. Bale, D. B. Betissimo, S. W. Grody, M. R. Hegde, E Lyon. B. E. Ward:
right determination with segmentation and anteroposterior axis elongation. Onobpment Molecular Subcommittee of the ACMG Laboratory Quality Assurance Committee. ACMG
132, S425-5436 12005). recommendations for standards for interpretation and reporting of sequence variations:
66. A. Tomita.Mitceell, O.K. Mahnke, C A. Strobl°, M. E. Turk-lett K. D. SIMIVI1, M. Hidestrand, Revisions 2007. Genet. Med. 10, 294-300 12008).
S. E. Harris. M. A. Goebel% P. M. Simpson, D. P. Erick, U. Stoeckel, A. N. Pdech. J. S. TweddelL
M. E. Mitchell, Human gene copy number spectra andpis in congenital heart malforma. Acknowledgments: We thank R. Cohen, V. Corkin. G. Richards. and &Bale for helpful comments.
lions. Physic( Gerson*, 44.518-541 (2012). as well as & Keithly and M. Clifton for their technical help. We thank P. Suter; H. Gudtjartsson.
67. S. C. Greenway. A. C. Pereira, J. C Lin. S. R. DePalma, S. J. brad, S. M. Mesquite, S. A Gohonsson. and K Stefannon from deCODE Genetic for assistance with sequence analysis.
E Ergots J. H. Conte. I M Kern, S. A. McCarrdl. J. M. Gorham. S. Gabriel. D. M. Albhuler, A deo AJOICI7. orb with again Finding This work was supported by the Marion March Dow
L Quintanteffieck Mde, M. A. Artunduaga. R. D. Eavey, K M. Plenge, N. A. Shadkk, Foundation Chicken', Men' Knead, and Mumma Inc. Author contribugons:05., SES.
M. E Wdntian, P. L De Jager, 0. A Haller. R. E Breitban, J. G. Seidman C. E. Seidman. and Sundertook analysis of data and confirmatory studies and helped write the manuscript
De novo copy number variants dandy new genes and loci in isolated sporadc tetralogy N. developed the tool for identification of variants that dint splicing M. compiled pa-
of Fella Abt. Cerra 41, 931-935 00093 tient information and assisted in data analysis:M. and= performed histcpathologrIM.
69, J. Christiansen, J. D. Dyck, B. G. Elyas, M. Lilley, J. S. Bardorth. M. Hicks, K. A. Sprysak, LAIC, and EGF. undertock panel, gimme. and dideoxy sequencing and data analysis; AF, Sic M.
R. Tomaszewski, S. M. Haase, L M. Vicen-Wyhony. M. J. Somerville, Chromosome 1821.1 1K. LOW, it RJCs. MI, and In developed the jI.Seq 2500 and undertook the geronric
contiguous gene deletion is associated with congenital heart disease. OK Res 94,1429-1435 stemming on that instrument M. had the ides to provide nee-generation sequences for ill
(20041. neonates; M. was the neonatologist of record for the patients; SF.K oversaw the work and
69. T. G. Willis, I. R. Zalcberg, L J. Coignet, I. Wlodarska, M. Slut, D. M. Jadayel, C. Bastard, wrote the manuscript Competing Interests: IL 511. LK, ACW.,1B. RJG, S.
and
J. G. Treleaven, 0. Catovsky, M. L Silva, M. J. Dyer. Molecular cloning of translocation Mare employees of Illumine Inc, whkh manufactures the Ildeq 2500 instrument. The other
01;101(q21;q32) defines a novel gene (DLLS) at chromosome 1q21. abed 91. 1873-1881 authors declare that they have no competing interests. Data and materials: The genomic
0993). sequence data for this study have been deposited in the database dbGAP. Please contact it
70. X. Yi C P. Ng. H. Habacher, 5. Roy, Fccirj1 transcription factors are master regulators of the authors for accession numbers.
motile ciliogenic program. Nat. Genet. 40, 1445-1453 (2004).
71. 1000 Genome Project Consortium, A map of human genorne variation from population. Submitted 19 March 2012
scale sequencing. Nature 467, 1061-1073 (2010). Accepted 4 September 2012
72. S. 5. Ajay. S. C. Parker, H. 0. Abaan, K. V. Fajardo, E. H. Margulies, Accurate and Published 3 October 2012 01
comprehensive sequencing of personal genomes. Genome Res. 21. 1498-1SOS 10.1126/scitranslmed.3004041 2
(2011). .0
73. 1. Sampson, K Jacobs, M. Yeager. S. Chinook, N. Chatterjee. Efficient study design for next Citation: Cf. Saunders. N. A.Miller, S. E. Soden. D. L.Dirmiddie. A.Not N. A. Alnacli.N. Andrews, LL
generation sequencing Genet. Epidemic,. 35, 269-277 (2011). L Penance; LA Krivohlavek, J.Fellis,S.Humphray,P.Saffrey,L Kngsbury,J.0 Weir. J. Betley. C
74. E R. Mardis, The 51,000 gencme, the 5100.000 analysis? Genome Med 2, 84 (2010). FI.J.GrocockE.IlManadies,EG.Fartow,M.Affinart.N.P.Safins.J.EPetrilin,KP.ItalLS.F.Kingsmcre. °
75. M. Q. 2hars. Statistical features of human exons and their Raiding regions. Num. Mot Rapid wholegenome sequencing for genetic disease diagnosis in neonatal intensive care units. ?
Genet 7, 919-932 (1998). Set Trent Med. 4. 154ra135 (2012).
tilt
E
C
0)
tq
E
E
co
co
3
www.ScienceTranslationalMedicine.org 3 October 2012 Vol 4 Issue 154 154ra135 13
EFTA00315113